Far infrared intelligent Dan moxibustion cooperative system based on big data analysis and intelligent Dan moxibustion bed

Abstract

The invention discloses a far-infrared intelligent Dan moxibustion cooperative system based on big data analysis and an intelligent Dan moxibustion bed, and the method comprises the steps: firstly, generating a homogenized acupoint distribution map through heat radiation scanning, combining an analysis algorithm with a physiological model for comparison, precisely marking a thermal anomaly region, and extracting a treatment point coordinate set; secondly, dynamically calculating a temperature gradient curve based on priority ranking and sensor feedback, and optimizing treatment parameters; and finally, fusing distance sensor data and a posture calibration system, determining an accurate treatment distance, and integrating duration control and vector data to form a continuous collaborative treatment sequence, so as to ensure treatment precision and adaptability. According to the method, through integrated innovation of heat energy distribution analysis, dynamic parameter adjustment and posture calibration, the accuracy and personalized adaptive capacity of acupoint treatment are remarkably improved, and an efficient solution is provided for accurate intervention of heat energy abnormity of a human physiological system.

Description

Technical Field

[0001] This invention relates to the field of moxibustion control technology, and in particular discloses a far-infrared intelligent moxibustion-drumming collaborative system and an intelligent moxibustion-drumming bed based on big data analysis. Background Technology

[0002] Traditional Chinese medicine moxibustion, as a time-honored method of health preservation and treatment, has unique value in regulating meridians, relieving chronic pain, and improving physical condition. Its application scope covers rheumatism, cervical and lumbar discomfort, as well as daily conditioning of sub-health conditions. Therefore, it still occupies an important position in modern health management.

[0003] In current moxibustion practice, the location of acupoints is primarily determined by the practitioner through touch, observation, and years of experience. The intensity of the moxa, the distance between the moxa and the skin, and the duration of treatment at each acupoint are also largely decided and adjusted on the spot. This reliance on personal experience leads to significant differences in the location and parameter selection for the same patient and symptoms among different doctors or technicians. Even when the same doctor performs the treatment at different times and under different fatigue levels, the intensity and precision of the moxibustion are difficult to maintain consistently, resulting in variations in the actual stimulation intensity received by the patient. This severely limits the repeatability and comparability of the treatment process.

[0004] A deeper problem lies in the fact that traditional methods struggle to provide clear and intuitive evidence before moxibustion to determine whether acupoints are truly in an active state "requiring stimulation," and to assess the overall blockage or imbalance of the meridians. Doctors can only indirectly infer this from the patient's description of symptoms, the soreness or distension felt upon pressure, or the local skin color. These subjective feelings are easily influenced by the patient's physical condition, emotions, and even the weather that day, leading to discrepancies between acupoint selection and the actual pathological state. When the wrong or unsuitable acupoints are selected, the subsequent energy input for moxibustion cannot accurately target the key areas of the lesion, thus reducing the overall effectiveness and relevance of the treatment.

[0005] Therefore, accurately capturing the true thermal characteristics of the body's meridians and acupoints before moxibustion begins, and using this as a basis to achieve personalized and precise matching of stimulation intensity, distance, and time for each acupoint, has become a key issue in ensuring the stable and consistent efficacy of moxibustion. Summary of the Invention

[0006] This invention provides a far-infrared intelligent moxibustion collaboration system and an intelligent moxibustion bed based on big data analysis, aiming to solve at least one of the defects existing in the above-mentioned prior art.

[0007] One aspect of the present invention relates to a far-infrared intelligent moxibustion collaborative system based on big data analysis, comprising:

[0008] The homogenized acupoint distribution map generation module is used to perform thermal radiation scanning on acupoints on the human body surface using a radiation detection device, collect physiological thermal data, and generate a homogenized acupoint distribution map through smoothing processing. The homogenized acupoint distribution map reflects the thermal energy state of the human physiological system.

[0009] The abnormal area marking module is used to compare the uniform acupoint distribution map with the preset physiological model template using an analysis algorithm to detect thermal abnormal deviations. When the thermal abnormal deviation exceeds the range, the abnormal area is marked.

[0010] The priority sorting module is used to analyze the thermal gradient changes in abnormal areas using a pattern matching algorithm, extract the coordinate set of acupoint treatment points and sort them according to priority, taking into account the severity of the abnormality.

[0011] The optimized temperature value acquisition module is used to acquire the priority sequence of the acupoint treatment point coordinate set, and is linked with the sensor feedback of the thermotherapy execution unit to calculate the temperature gradient curve for dynamically adjusting treatment parameters and obtaining the optimized temperature value.

[0012] The precise treatment distance parameter determination module for moxibustion is used to fuse and optimize temperature values ​​and distance sensor data, assess the impact of posture deviation, and determine the precise treatment distance parameter for moxibustion by calibrating the positioning system if posture deviation causes a deviation.

[0013] The integrated operation execution module is used to integrate the vector data of the precise treatment distance parameters, duration control logic, and posture adaptation mechanism of Danzhi to form a continuous collaborative treatment sequence and execute integrated operation to ensure treatment accuracy and adaptability.

[0014] Furthermore, the module for generating a homogenized acupoint distribution map includes:

[0015] The two-dimensional scanning matrix conversion unit is used to acquire the radiation intensity signal collected by the radiation detection device and convert the radiation intensity signal into a two-dimensional scanning matrix according to the acupoint coordinate mapping relationship.

[0016] The continuous gradient distribution model generation unit is used to perform median filtering and bicubic interpolation calculations on the two-dimensional scan matrix to generate a continuous gradient distribution model.

[0017] The homogenized acupoint distribution map generation unit is used to process the continuous gradient distribution model using pixel mapping technology to generate a homogenized acupoint distribution map. The contrast of the homogenized acupoint distribution map is enhanced by histogram equalization.

[0018] The thermal energy state determination unit is used to extract the state characteristics of the homogenized acupoint distribution map and determine the thermal energy state of the human physiological system based on the state characteristics.

[0019] Furthermore, the abnormal region marking module includes:

[0020] The standard reference thermal field generation unit is used to obtain a homogenized acupoint distribution map and load a preset physiological model template to generate a standard reference thermal field with spatial alignment.

[0021] The gray-scale difference matrix construction unit is used to perform differential calculations between the homogenized acupoint distribution map and the standard reference thermal field to construct the gray-scale difference matrix;

[0022] The thermal anomaly deviation distribution set generation unit is used to weight the gray-level difference matrix by combining the regional sensitivity coefficient to generate the thermal anomaly deviation distribution set.

[0023] An abnormal feature point cloud set extraction unit is used to extract an abnormal feature point cloud set if the thermal anomaly deviation distribution set exceeds a preset threshold.

[0024] Anomaly region marking unit is used to process the set of anomalous feature point clouds to extract edge contours and mark anomalous regions on the homogenized acupoint distribution map.

[0025] Furthermore, the priority sorting module includes:

[0026] The local thermal gradient vector field calculation unit is used to calculate the local thermal gradient vector field of the pixel grayscale matrix within the abnormal region. The local thermal gradient vector field reflects the rate of change of the direction of the pixel.

[0027] The candidate treatment point center coordinate localization unit is used to perform normalized cross-correlation matching between the local thermal gradient vector field and the pre-set acupoint thermal feature template library to locate the center coordinates of the candidate treatment point.

[0028] The quantitative anomaly severity value acquisition unit is used to acquire the quantitative anomaly severity value corresponding to the center coordinates of the candidate treatment point. The quantitative anomaly severity value represents the degree of deviation of the gradient magnitude from the preset standard physiological model.

[0029] The acupoint treatment point coordinate set generation unit is used to sort the center coordinates of candidate treatment points in descending order based on the quantified severity of abnormality values, and generate a set of acupoint treatment point coordinates sorted by priority.

[0030] Furthermore, the temperature value acquisition module has been optimized, including:

[0031] The priority sequence generation unit is used to obtain the coordinate set of acupoint treatment points and calculate the thermal correlation degree to generate a priority sequence of the acupoint treatment point coordinate set.

[0032] The sensor feedback data stream acquisition unit is used to control the hyperthermia execution unit according to the priority sequence and collect thermal radiation signals to obtain the sensor feedback data stream;

[0033] The dynamic adjustment treatment parameter generation unit is used to fit the temperature gradient curve based on the sensor feedback data stream. If the rate of change of the temperature gradient curve exceeds the threshold, dynamic adjustment treatment parameters are generated.

[0034] The optimized temperature value acquisition unit is used to feed back dynamically adjusted treatment parameters to the thermotherapy execution unit for iterative approximation to obtain the optimized temperature value.

[0035] Furthermore, the module for determining the precise treatment distance parameters for moxibustion includes:

[0036] The associated dataset establishment unit is used to receive optimized temperature values ​​and trigger the distance sensor to collect real-time physical distance data to establish an associated dataset of temperature and distance.

[0037] The three-dimensional space calibration instruction generation unit is used to call the positioning system to obtain spatial coordinates and calculate the attitude offset vector based on the associated dataset. If the attitude offset vector is greater than a preset threshold, a three-dimensional space calibration instruction is generated.

[0038] The real-time physical distance data acquisition unit is used to drive the execution unit to complete the pose correction according to the three-dimensional space calibration command and acquire the corrected real-time physical distance data.

[0039] The Danzhi Precision Treatment Distance Parameter Output Unit is used to fuse optimized temperature values ​​and corrected real-time physical distance data, calculate and output the Danzhi precision treatment distance parameters.

[0040] Furthermore, the integrated operation execution module includes:

[0041] The continuous collaborative treatment sequence generation unit is used to acquire the precise treatment distance parameters, duration control logic, and vector data of the posture adaptation mechanism of Danzhi, and to map the vector data of the posture adaptation mechanism to generate a continuous collaborative treatment sequence.

[0042] The actual motion trajectory model construction unit is used to analyze the continuous collaborative treatment sequence to obtain the underlying pulse control signal, drive the posture adaptation mechanism to operate based on the underlying pulse control signal, and construct the actual motion trajectory model.

[0043] The correction instruction execution unit is used to calculate the deviation matrix between the actual motion trajectory model and the continuous synergistic treatment sequence, generate a compensation vector based on the deviation matrix, and execute correction instructions to ensure treatment accuracy and adaptability.

[0044] Another aspect of the present invention relates to an intelligent moxibustion bed, comprising an intelligent bed body and a far-infrared intelligent moxibustion coordination system based on big data analysis, as shown above, disposed on the intelligent bed body.

[0045] The beneficial effects achieved by this invention are as follows:

[0046] This invention provides a far-infrared intelligent moxibustion collaborative system and an intelligent moxibustion bed based on big data analysis. Addressing the core business scenario issues of insufficient treatment precision due to uneven distribution of physiological thermal energy in the human body, and the impact of posture deviation on treatment efficacy, it achieves precise treatment through multi-dimensional data fusion and dynamic calibration mechanisms. First, a uniform acupoint distribution map is generated through thermal radiation scanning. Combined with analysis algorithms and comparison with physiological models, thermally abnormal areas are accurately marked and the coordinate set of treatment points is extracted. Second, based on priority ranking and sensor feedback, the temperature gradient curve is dynamically calculated to optimize treatment parameters. Finally, distance sensor data and a posture calibration system are fused to determine the precise treatment distance, and duration control and vector data are integrated to form a continuous collaborative treatment sequence, ensuring treatment accuracy and adaptability. The overall technical effect of this invention lies in its integrated innovation of thermal energy distribution analysis, dynamic parameter adjustment, and posture calibration, significantly improving the accuracy and personalized adaptability of acupoint treatment, providing an efficient solution for precise intervention of abnormal thermal energy in the human physiological system. Attached Figure Description

[0047] Figure 1 This is a functional block diagram of an embodiment of a far-infrared intelligent moxibustion collaborative system based on big data analysis according to the present invention.

[0048] Explanation of icon numbers:

[0049] 10. Module for generating a uniform acupoint distribution map; 20. Module for marking abnormal areas; 30. Module for prioritizing and sorting; 40. Module for optimizing temperature value acquisition; 50. Module for determining the precise treatment distance parameter of moxibustion; 60. Module for integrated operation and execution. Detailed Implementation

[0050] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0051] like Figure 1As shown, the first embodiment of the present invention proposes a far-infrared intelligent moxibustion collaborative system based on big data analysis, including a homogenized acupoint distribution map generation module 10, an abnormal area marking module 20, a priority sorting module 30, an optimized temperature value acquisition module 40, a moxibustion precise treatment distance parameter determination module 50, and an integrated operation execution module 60. The homogenized acupoint distribution map generation module 10 is used to perform thermal radiation scanning on acupoints on the human body surface using a radiation detection device, collect physiological thermal data, and generate a homogenized acupoint distribution map through smoothing processing. The homogenized acupoint distribution map reflects the thermal energy state of the human physiological system. The abnormal area marking module 20 is used to compare the homogenized acupoint distribution map with a preset physiological model template using an analysis algorithm to detect thermal anomalies. When the thermal anomaly deviation exceeds the range, an abnormal area is marked. The priority sorting module 30... 0. The module is used to analyze the thermal gradient changes in abnormal areas using a pattern matching algorithm, extract the coordinate set of acupoint treatment points, and sort them according to priority, taking into account the severity of the abnormality. The optimized temperature value acquisition module 40 is used to acquire the priority sequence of the acupoint treatment point coordinate set, link with the sensor feedback of the thermotherapy execution unit, calculate the temperature gradient curve, and dynamically adjust the treatment parameters to obtain the optimized temperature value. The precise treatment distance parameter determination module 50 is used to fuse the optimized temperature value and distance sensor data, evaluate the impact of posture deviation, and determine the precise treatment distance parameter of the thermotherapy by calibrating through the positioning system if the posture deviation causes a deviation. The integrated operation execution module 60 is used to integrate the precise treatment distance parameter of the thermotherapy, the duration control logic, and the vector data of the posture adaptation mechanism to form a continuous collaborative treatment sequence and execute integrated operation to ensure treatment accuracy and adaptability.

[0052] The homogenized acupoint distribution map generation module 10 uses a dedicated radiation detection device to perform non-contact thermal radiation scanning on the meridians and acupoints on the human body surface, accurately capturing the physiological thermal data of each acupoint and surrounding tissues; then, through smoothing and noise reduction processing to eliminate environmental interference and equipment acquisition errors, a homogenized acupoint distribution map is finally generated. This homogenized acupoint distribution map can intuitively and accurately reflect the thermal energy distribution state of the human physiological system, providing standardized basic data support for the subsequent abnormal area marking module 20 to detect abnormalities. It is the source link to ensure the reliability of the "detection-treatment" collaboration.

[0053] The abnormal area marking module 20 uses the homogenized acupoint distribution map generated by the homogenized acupoint distribution map generation module 10 as its core input. Through the application of professional analysis algorithms, it precisely compares the thermal energy distribution data of each acupoint and surrounding tissue in the map with a preset physiological model template (including the normal human meridian thermal energy distribution benchmark and the standard range of acupoint thermal gradient), point by point and region by region, to quantitatively detect thermal anomalies in each region. Simultaneously, it sets a scientific deviation range threshold; when the thermal anomaly deviation of a certain region exceeds this range, it is automatically marked as an abnormal region. Its core function is to receive the basic detection data from the homogenized acupoint distribution map generation module 10, completing the transformation from "thermal energy state presentation" to "abnormal information extraction." It provides precise target areas for the thermal gradient analysis and treatment point extraction of the priority ranking module 30, and provides core abnormal data support for subsequent treatment parameter adjustment by the temperature value acquisition module 40, distance calibration by the moxibustion precise treatment distance parameter determination module 50, and the formation of continuous collaborative treatment sequences by the integrated operation execution module 60. It is a key hub ensuring a collaborative closed loop of "detection-treatment."

[0054] The priority ranking module 30 uses the abnormal area marked by the abnormal area marking module 20 as the analysis scope, and uses a pattern matching algorithm to analyze the thermal gradient change pattern within the area, accurately extracting the coordinate set of acupoint treatment points that are strongly correlated with the abnormal thermal energy distribution. At the same time, it sets priority ranking rules based on the degree of thermal anomaly deviation quantified by the abnormal area marking module 20. The more severe the anomaly, the higher the ranking of the treatment point. This provides a targeted treatment point sequence for the subsequent optimization of the temperature value acquisition module 40 to accurately match treatment parameters, ensuring that treatment resources are tilted towards the core abnormal area.

[0055] The optimized temperature value acquisition module 40 integrates the priority sequence of acupoint treatment points output by the priority sorting module 30. Through real-time sensor feedback data from the thermotherapy execution unit and linkage with the treatment point sequence, the temperature gradient curve of each treatment point is calculated using a heat conduction simulation algorithm. Based on the curve characteristics, the core parameters of moxibustion treatment (temperature, power, and action rhythm) are dynamically adjusted, and finally, the optimized temperature value that matches the priority of each treatment point is determined, achieving a precise match of "targeted treatment point + personalized temperature" to avoid tissue damage from excessively high treatment temperature or efficacy impact from excessively low treatment temperature.

[0056] The precise treatment distance parameter determination module 50 of Danzhi integrates the optimized temperature value of the optimized temperature value acquisition module 40 with the real-time data collected by the distance sensor to construct a "temperature-distance" adaptation model. This model assesses the impact of human posture deviation on treatment distance and temperature conduction during treatment. If posture deviation causes the treatment distance deviation to exceed the allowable range, the position of the Danzhi execution mechanism is calibrated in real time through the positioning system. Ultimately, the precise treatment distance parameters of each treatment point are determined to ensure that the optimized temperature value can accurately act on the target acupoints and avoid treatment deviation caused by posture deviation.

[0057] The integrated operation execution module 60 integrates the precise treatment distance parameters of the moxibustion precision treatment distance parameter determination module 50, the duration control logic based on treatment point priority (the higher the priority, the longer the treatment time), and the vector data of the posture adaptation mechanism. Through a collaborative scheduling algorithm, it constructs a standardized continuous collaborative treatment sequence; drives the moxibustion execution mechanism, positioning mechanism, and posture adaptation mechanism to operate in an integrated manner according to the sequence, ensuring precise positioning of treatment points, stable and controllable temperature, and dynamic posture adaptation throughout the process. Ultimately, it achieves precise and automated collaboration of the entire "detection-treatment" process, ensuring treatment accuracy and human adaptability.

[0058] Furthermore, the far-infrared intelligent moxibustion collaborative system based on big data analysis provided in this embodiment includes a homogenized acupoint distribution map generation module 10 comprising a two-dimensional scanning matrix conversion unit, a continuous gradient distribution model generation unit, a homogenized acupoint distribution map generation unit, and a thermal energy state determination unit. The two-dimensional scanning matrix conversion unit is used to acquire the radiation intensity signal collected by the radiation detection device and convert the radiation intensity signal into a two-dimensional scanning matrix according to the acupoint coordinate mapping relationship.

[0059] The two-dimensional scan matrix is ​​obtained through the following formula:

[0060] (1)

[0061] In formula (1), Indicates the two-dimensional scan matrix in The pixel value of the location, The number of radiation samples collected by the radiation detection device is... One radiation intensity signal, The function represents the mapping relationship between acupoint coordinates, which maps the matrix positions. Mapped to the corresponding signal index The control logic of formula (1) takes the radiation intensity signal and the acupoint coordinate mapping function as inputs. Through the process of "matrix position-signal index binding → position-by-position signal assignment → complete matrix construction", the one-dimensional radiation detection signal is converted into a two-dimensional scanning matrix that fits the spatial distribution of acupoints. The core is to establish the correspondence between the matrix pixels and the radiation signals of the acupoint area by means of the mapping function, so as to provide basic data for the subsequent generation of a uniform acupoint distribution map.

[0062] The two-dimensional scanning matrix conversion unit first processes the radiation intensity signals acquired by radiation detection devices such as infrared thermal imagers. These signals are typically represented digitally as the thermal radiation levels of different parts of the human body. Based on the acupoint coordinate mapping relationship, these radiation intensity signals are converted into a two-dimensional scanning matrix, mapping the scattered data onto a grid structure. Specifically, assuming the detection device scans the front of the human body and acquires radiation values ​​for acupoints related to the Lung Meridian of Hand-Taiyin, such as 45 units at the Feishu acupoint and 50 units at the Tanzhong acupoint, these points are converted into a 100×100 matrix using a predefined coordinate system, such as mapping the center line of the human body as the Y-axis and the shoulder width as the X-axis. Each matrix element corresponds to the average radiation value of a pixel, thus forming the initial two-dimensional representation, which facilitates subsequent image processing.

[0063] The continuous gradient distribution model generation unit is used to perform median filtering and bicubic interpolation calculations on the two-dimensional scan matrix to generate a continuous gradient distribution model.

[0064] The two-dimensional scan matrix after median filtering is obtained by the following formula:

[0065] (2)

[0066] In formula (2), Indicates the median filter position. The pixel value of the location, express All within the center's neighborhood Take the median. The control logic of formula (2) uses the pixel neighborhood of the two-dimensional scanning matrix as the processing unit. Through the process of "neighborhood selection → pixel value sorting → median assignment", it filters out abnormal pixel values ​​caused by radiation detection noise in the matrix and retains the true gradient characteristics of acupoint radiation intensity. The core is to use the characteristic that the median is not sensitive to extreme values ​​to eliminate detection interference and lay a clean foundation for the subsequent generation of continuous gradient distribution model.

[0067] After bicubic interpolation, the image is at continuous positions. The grayscale value at a given location is obtained using the following formula:

[0068] (3)

[0069] In formula (3), This indicates that the image at continuous positions after bicubic interpolation grayscale value at the location, continuous coordinates The smoothed grayscale value (or acupoint radiation intensity value) is the final output. and This indicates the basis function of the bicubic interpolation in the corresponding fractional part. and The values ​​on the basis function are used to calculate the weights of neighboring pixels, and the smoothness of the basis function ensures the continuity of the interpolation results. Represents the discrete pixel values ​​of a 4×4 neighborhood, using continuous coordinates. integer part Based on this, a 4×4 neighborhood pixel matrix was selected. , is the input basis for interpolation. The control logic of formula (3) takes the discrete pixel matrix and the fractional part of the continuous position as input. Through the process of "position decomposition → neighborhood selection → basis function weighting → multi-pixel summation", the discrete two-dimensional scanning matrix is ​​converted into a continuous and smooth gray-level distribution, which provides high-resolution continuous data for the subsequent generation of continuous gradient distribution model. The core is to use the smooth weighting characteristics of bicubic basis functions to realize the natural transition from discrete pixels to continuous gradients and preserve the detailed gradient of acupoint radiation intensity.

[0070] The following formula is used to construct a continuous gradient intensity distribution model:

[0071] (4)

[0072] In formula (4), This indicates that the continuous gradient distribution model is at the location The intensity value at that location, This represents the gradient vector of the interpolation function. and These represent the images after bicubic interpolation at... direction and The partial derivatives in the direction. The control logic of formula (4) is based on the continuous gray distribution after bicubic interpolation. Through the process of "partial derivative calculation → gradient vector magnitude solution → continuous distribution construction", the degree of change of radiation intensity in the acupoint area is quantified, and a continuous distribution model reflecting the radiation gradient is generated. The core is to use the gradient magnitude to reflect the local change characteristics of radiation intensity, and to provide gradient basis for the subsequent homogenization of acupoint distribution.

[0073] The continuous gradient distribution model generation unit performs median filtering on this two-dimensional scan matrix to remove noise. For example, if there are abnormally high values ​​in the two-dimensional scan matrix due to environmental interference, such as a pixel with a value of 80 while the surrounding pixels are all 40, median filtering through a 3×3 window replaces it with the median value of 45 within the window, thus smoothing the data. Next, bicubic interpolation is performed, a higher-order interpolation method. The principle is to use the weighted average of the surrounding 16 pixels to estimate the new point value, generating a continuous gradient distribution model. For example, when interpolating in the blank areas of the two-dimensional scan matrix, although the calculation formula does not need to be detailed, the process involves polynomial fitting to ensure that the gradient smoothly transitions from high-radiation areas such as the abdomen to low-radiation areas such as the limbs. Finally, the continuous gradient distribution model presents a continuous change map of human body thermal energy; for example, the heart region shows a red gradient to indicate high heat.

[0074] The homogenized acupoint distribution map generation unit is used to process the continuous gradient distribution model using pixel mapping technology to generate a homogenized acupoint distribution map. The contrast of the homogenized acupoint distribution map is enhanced by histogram equalization.

[0075] The homogenized acupoint distribution map is obtained using the following formula:

[0076] (5)

[0077] In formula (5), This indicates the pixel position of the homogenized acupoint distribution map. grayscale value at that location Represents pixels The corresponding continuous space mapping bucket, The size of the mapping bucket is indicated. The control logic of formula (5) is based on the continuous gradient distribution model and the pixel mapping bucket. Through the process of "continuous space-pixel grid mapping → gradient aggregation within the bucket → mean assignment", the continuous gradient distribution is converted into a uniform acupoint distribution map of uniform pixel grid, realizing the uniform discrete expression of acupoint radiation gradient. The core is to use the "mapping bucket" to associate the continuous space and uniform pixels, and smooth the gradient through the mean within the bucket, laying the foundation for uniform distribution for subsequent contrast enhancement.

[0078] The homogenized acupoint distribution map generation unit uses pixel mapping technology to process the continuous gradient distribution model, normalizing the gradient values ​​to the range of 0-255 to generate a homogenized acupoint distribution map. Here, homogenization means adjusting the uneven distribution to ensure that the heat values ​​of acupoints such as Hegu (LI4) are displayed evenly in the homogenized acupoint distribution map, avoiding local overexposure or underexposure. Subsequently, histogram equalization enhances contrast. This works by stretching the pixel grayscale distribution; for example, if the original histogram is concentrated in the middle value, it is remapped using a cumulative distribution function, making low-contrast areas such as leg acupoints clearer. This enhances the visual and analytical effects of the map, thus better highlighting abnormal heat areas in medical diagnosis.

[0079] The thermal energy state determination unit is used to extract the state characteristics of the homogenized acupoint distribution map and determine the thermal energy state of the human physiological system based on the state characteristics.

[0080] The thermal energy state of the human physiological system is obtained by the following formula:

[0081] (6)

[0082] In formula (6), This indicates the thermal energy state of the human physiological system. The number of pixel positions is indicated. The control logic of formula (6) is based on the continuous gradient distribution model and pixel mapping bucket. Through the process of "continuous space-pixel grid mapping → gradient aggregation within the bucket → mean assignment", the continuous gradient distribution is converted into a uniform pixel grid map, realizing the uniform discrete expression of the acupoint radiation gradient. The core is to use the "mapping bucket" to associate the continuous space with uniform pixels, and smooth the gradient through the mean within the bucket, laying the foundation for uniform distribution for subsequent contrast enhancement.

[0083] The thermal state determination unit extracts state characteristics from the homogenized acupoint distribution map, involving calculations of indicators such as average calorific value and gradient variance. For example, the global average for the entire map is 48 units, while the variance for a local area, such as the Liver Shu acupoint, is 15, indicating large thermal energy fluctuations. Based on these state characteristics of the homogenized acupoint distribution map, the thermal state of the human physiological system is determined. For instance, if the average calorific value of acupoints related to the Lung Meridian exceeds the threshold of 55 and is accompanied by high variance, it is inferred that there is excessive thermal energy in the respiratory system, suggesting an inflammatory risk. This method can provide quantitative evidence in TCM thermal imaging diagnosis, improving the accuracy of the transition from radiation signals to physiological assessment.

[0084] Furthermore, the far-infrared intelligent moxibustion collaborative system based on big data analysis provided in this embodiment includes an abnormal region marking module 20 comprising a standard reference thermal field generation unit, a gray-scale difference matrix construction unit, a thermal anomaly deviation distribution set generation unit, an abnormal feature point cloud set extraction unit, and an abnormal region marking unit. The standard reference thermal field generation unit is used to obtain a homogenized acupoint distribution map and load a preset physiological model template to generate a standard reference thermal field with spatially aligned positions.

[0085] The standard reference thermal field for spatial alignment is derived using the following formula:

[0086] (7)

[0087] In formula (7), Indicates the standard reference thermal field at the pixel location. grayscale value at that location Indicates the spatial pixel position. Indicates the number of acupoints. Indicates the spatial pixel position The corresponding acupoint positions, Indicates the weighting coefficient. The thermal field diffusion parameter is represented. The control logic of formula (7) is based on the standard acupoints of the preset physiological model. Through the process of "spatial position alignment → single acupoint thermal field diffusion calculation → multi-acupoint contribution aggregation", a normal acupoint thermal field model matching the current detection map space is generated. The core is to use Gaussian diffusion to simulate the thermal distribution characteristics of normal acupoints, and combine acupoint weights to reflect the standard thermal field characteristics of the physiological model, so as to provide a reference benchmark for subsequent abnormal detection.

[0088] The process of acquiring a homogenized acupoint distribution map by the standard reference thermal field generation unit first involves reading data from the previous smoothing results. For example, in a traditional Chinese medicine thermal imaging system, the scanned map is stored as an image file. The system automatically loads it and calls a preset physiological model template. This physiological model template is a standard thermal distribution model built based on a large amount of data from healthy individuals, containing the average thermal coordinates of major acupoints such as Taiyuan and Shenmen. When generating a spatially aligned standard reference thermal field, the homogenized acupoint distribution map needs to be matched with the physiological model template in terms of coordinates. For example, affine transformation methods can be used to adjust the scale to ensure that acupoints such as Quchi coincide in both, thus forming an aligned thermal field image for subsequent comparison.

[0089] The grayscale difference matrix construction unit is used to perform differential calculations between the homogenized acupoint distribution map and the standard reference thermal field to construct the grayscale difference matrix.

[0090] The gray-level difference matrix is ​​obtained using the following formula:

[0091] (8)

[0092] In formula (8), This indicates the position of the gray-level difference matrix. The element at that location, This indicates the pixel position of the homogenized acupoint distribution map. The grayscale value. The control logic of formula (8) takes the actual thermal field and the standard thermal field aligned in space as input, and through the process of "one-to-one matching of positions → pixel-by-pixel difference calculation → difference matrix construction", it quantifies the degree of deviation between the actual acupoint thermal field and the normal physiological thermal field, and provides a basis for deviation data for subsequent identification of abnormal areas. The core is to use the difference between "actual and standard" to directly reflect the abnormal deviation of the thermal field. The magnitude / positive and negative of the difference corresponds to the degree of abnormality.

[0093] When the grayscale difference matrix construction unit performs difference calculations on the homogenized acupoint distribution map and the standard reference thermal field, it performs pixel-level subtraction operations. Specifically, it first converts both into grayscale matrices of the same resolution, such as 256×256, and then subtracts them point by point to obtain the difference. For example, if the grayscale of a certain point in the homogenized acupoint distribution map is 60 and the reference thermal field is 50, the difference is 10. The grayscale difference matrix constructed in this way reflects the deviation area of ​​the thermal distribution.

[0094] The thermal anomaly deviation distribution set generation unit is used to weight the gray-level difference matrix by combining the regional sensitivity coefficient to generate the thermal anomaly deviation distribution set.

[0095] The following formula is used to generate the elements of the thermal anomaly deviation distribution set by multiplying the gray-level difference matrix by region weighting:

[0096] (9)

[0097] In formula (9), Elements representing the distribution set of thermal anomaly deviations. The element representing the regional sensitivity coefficient. The control logic of formula (9) takes the gray-scale difference matrix and the regional sensitivity coefficient as inputs, and through the process of "one-to-one location matching → weighted amplification / weakening of deviation in each region → construction of distribution set", it makes the deviation data focus on the key acupoint area in physiological sense, strengthens the abnormal deviation of important areas, weakens the interference of secondary areas, and generates a more targeted thermal abnormal deviation distribution set. The core is to use the regional sensitivity coefficient to reflect the physiological importance of acupoint area and improve the accuracy of abnormal detection.

[0098] The thermal anomaly deviation distribution set generation unit combines regional sensitivity coefficients to weight the gray-level difference matrix. This involves introducing a coefficient table, which sets values ​​based on the sensitivity of human body parts such as the chest and abdomen. For example, the coefficient for the liver area is 1.5 while that for the limbs is 1.0. After weighting, a thermal anomaly deviation distribution set is generated. For example, the differences are multiplied by coefficients and then summarized into a set to highlight anomalies in important areas.

[0099] The abnormal feature point cloud set extraction unit is used to extract the abnormal feature point cloud set if the thermal anomaly deviation distribution set exceeds a preset threshold.

[0100] The following formula defines the filtering rules for anomalous point clouds, used to determine whether the entire thermal anomaly deviation distribution set exceeds a preset threshold:

[0101] (10)

[0102] In formula (10), This represents the extracted set of anomalous feature points. Indicates the first Point cloud coordinates, Indicates the first Temperature deviation value of each point cloud coordinate point Indicates the mean deviation. Indicates the preset threshold. This represents the number of point clouds in the thermal anomaly deviation distribution set. The control logic of formula (10) is based on the point cloud data and the mean deviation of the thermal anomaly deviation distribution set. Through the process of "relative deviation calculation → threshold screening → abnormal point cloud aggregation", it accurately extracts the abnormal point clouds that deviate from the overall deviation level. The core is to combine the "mean deviation" to judge the excessive deviation of a single point, avoid misjudgment caused by the overall deviation shift, and ensure the rationality of the abnormal point clouds.

[0103] In the abnormal feature point cloud set extraction unit, if the thermal anomaly deviation distribution set exceeds a preset threshold, such as an average deviation greater than 20, the process of extracting the abnormal feature point cloud set is to filter high difference points from the matrix. For example, a threshold segmentation algorithm is used to isolate the point group with deviation greater than the threshold, forming a three-dimensional point cloud representation, where each point contains location and deviation intensity information.

[0104] Anomaly region marking unit is used to process the set of anomalous feature point clouds to extract edge contours and mark anomalous regions on the homogenized acupoint distribution map.

[0105] The following formula is used to mark abnormal areas on a homogenized acupoint distribution map:

[0106] (11)

[0107] In formula (11), Indicates the location on the homogenized acupoint distribution map. The abnormal label score, Point Projection onto a homogenized acupoint distribution map The width of the Gaussian kernel is represented. The control logic of formula (11) takes the abnormal feature point cloud and the pixel position of the map as input, and quantifies the degree of association between each position of the map and the abnormal point cloud through the process of "point cloud-map projection → local region association → Gaussian contribution calculation → maximum association assignment". It generates anomaly label scores to define the abnormal area. The core is to use the Gaussian function to reflect the spatial association of "near strong and far weak", take the maximum value to strengthen the local influence of the abnormal point, and accurately mark the spatial range of the abnormal area.

[0108] When processing a set of anomalous feature point clouds to extract edge contours, the abnormal region marking unit employs convex hull algorithms or edge tracking methods. For example, it calculates boundary points from the point cloud, connects them to form contour lines, and then overlays markings onto a homogenized acupoint distribution map. This includes using red outlines to mark acupoint groups corresponding to abnormal regions, which aids in visual diagnosis. Further, in practical applications, this marking could be integrated into a mobile app, allowing users to directly see prominent areas when viewing the map, thus facilitating the tracking of health changes.

[0109] Furthermore, the far-infrared intelligent moxibustion collaborative system based on big data analysis provided in this embodiment includes a priority sorting module 30 comprising a local thermal gradient vector field calculation unit, a candidate treatment point center coordinate positioning unit, a quantitative abnormality severity numerical acquisition unit, and an acupoint treatment point coordinate set generation unit. The local thermal gradient vector field calculation unit is used to calculate the local thermal gradient vector field of the pixel grayscale matrix within the abnormal region. The local thermal gradient vector field reflects the directional change rate of the pixel.

[0110] The local thermal gradient vector field of the pixel grayscale matrix within the abnormal region is obtained by the following formula:

[0111] (12)

[0112] In formula (12), Represents pixels The local thermal gradient vector at that location, This represents the rate of change of grayscale at that point in the horizontal direction. This represents the grayscale change rate in the vertical direction of the point. The control logic of formula (12) is based on the pixel grayscale distribution in the abnormal region. Through the process of "local pixel difference → direction change rate calculation → gradient vector combination", a vector field reflecting the direction and rate of thermal field change in the abnormal region is constructed, which provides the direction basis of thermal gradient for subsequent location of core treatment points. The core is to use the grayscale change rate in the horizontal and vertical directions to quantify the local thermal field change characteristics of each pixel.

[0113] The process of calculating the local thermal gradient vector field of the pixel grayscale matrix within an abnormal region by the local thermal gradient vector field calculation unit first requires extracting the corresponding pixel grayscale matrix from the previously marked abnormal region. For example, in a traditional Chinese medicine thermal imaging system, this pixel grayscale matrix is ​​a 128×128 two-dimensional array, where each element represents the thermal intensity value of an acupoint region, such as the Hegu acupoint. Next, the gradient is calculated by applying the Sobel operator. Each vector in the vector field consists of gradient components in the horizontal and vertical directions, reflecting the rate of directional change of the pixel's thermal value. For instance, if the grayscale value of a pixel rapidly increases from 40 on the left to 70 on the right, its vector will point to the right and have a larger amplitude. The resulting vector field can highlight the edges and directional changes in thermal distribution, helping to identify potential acupoint abnormal patterns.

[0114] The candidate treatment point center coordinate localization unit is used to perform normalized cross-correlation matching between the local thermal gradient vector field and the pre-set acupoint thermal feature template library to locate the center coordinates of the candidate treatment point.

[0115] The following formula is used to locate the center coordinates of the local thermal gradient vector field and the optimal template by maximizing the normalized cross-correlation, and these coordinates are used as candidate treatment points:

[0116] (13)

[0117] In formula (13), Indicates the center coordinates of the candidate treatment point. Indicates the candidate position. Represents the normalized cross-correlation function. The best-matching acupoint thermal feature template is represented. The control logic of formula (13) is based on the local thermal gradient vector field and the best acupoint thermal template. Through the process of "candidate position traversal → template-region similarity calculation → maximum matching degree positioning", the center position in the abnormal region that best matches the standard treatment feature is found as the candidate treatment point. The core is to use normalized cross-correlation (NCC) to measure the similarity between the region and the template, and maximize the matching degree to ensure that the candidate point fits the best treatment feature.

[0118] When the candidate treatment point center coordinate localization unit performs normalized cross-correlation matching between the local thermal gradient vector field and the pre-set acupoint thermal feature template library, the pre-set acupoint thermal feature template library is a collection built based on historical health data, containing typical gradient patterns of various standard acupoints such as Neiguan acupoint. Each template is a small-sized vector field matrix, such as 16×16, used to represent the thermal change characteristics under normal physiological conditions. The matching process involves a sliding window operation. Specifically, the normalized cross-correlation coefficient is calculated point-by-point between the vector field and each template. For example, a similarity score is calculated using a formula. When the score exceeds 0.8, it is marked as a matching point, thereby locating the center coordinates of the candidate treatment point. For example, a center point is found in the lung meridian-related area, corresponding to the treatment acupoint.

[0119] The quantitative anomaly severity value acquisition unit is used to acquire the quantitative anomaly severity value corresponding to the center coordinates of the candidate treatment point. The quantitative anomaly severity value represents the degree of deviation of the gradient magnitude from the preset standard physiological model.

[0120] The numerical value for quantifying the severity of anomalies is obtained using the following formula:

[0121] (14)

[0122] In formula (14), This indicates a numerical value representing the severity of the quantification anomaly. This represents the actual gradient magnitude. The control logic of formula (14) is based on the actual gradient and standard gradient of the candidate treatment point. Through the process of "gradient extraction at the corresponding position → amplitude difference calculation → deviation degree quantification", the difference between the actual and standard gradients is used to characterize the severity of the abnormality of the candidate treatment point. The core is to use the deviation of the gradient amplitude of "actual-standard" to intuitively reflect the degree of abnormality of the point relative to the normal physiological state.

[0123] The quantitative severity assessment unit obtains the quantitative severity value corresponding to the center coordinates of the candidate treatment point. This requires first reviewing the preset standard physiological model, a benchmark model statistically derived from a large number of samples, including the average gradient amplitude range of acupoints such as Zusanli (ST36). The quantification process calculates the difference between the Euclidean norm of the gradient vector at the candidate point and the corresponding value in the model. For example, if the model's standard amplitude is 15 while the actual amplitude is 25, the deviation is 10. Further normalization yields a value between 0 and 1, representing the severity, which facilitates subsequent ranking.

[0124] The acupoint treatment point coordinate set generation unit is used to sort the center coordinates of candidate treatment points in descending order based on the quantified severity of abnormality values, and generate a set of acupoint treatment point coordinates sorted by priority.

[0125] The following formula is used to generate a set of acupoint treatment point coordinates by selecting the corresponding center coordinates in descending order based on the numerical ranking of the severity of the quantified abnormality:

[0126] (15)

[0127] In formula (15), This represents a set of coordinates of acupoints for treatment, sorted by priority. This indicates the number of values ​​sorted in descending order of the severity of the quantitative anomaly. One value, Indicates the corresponding number The center coordinates of candidate treatment points for each severity level, The total number of candidate treatment points is represented. The control logic of formula (15) is based on the severity of the abnormality of the candidate treatment points. Through the process of "pairing information sorting → severity descending order sorting → priority coordinate set generation", the candidate treatment points are arranged from high to low according to the severity of the abnormality, and a treatment point coordinate set with clear priority is obtained. The core is to arrange the treatment points with more severe abnormalities in descending order so that they are treated first, thus ensuring the pertinence and efficiency of moxibustion treatment.

[0128] When the acupoint treatment point coordinate set generation unit sorts the center coordinates of candidate treatment points in descending order based on the quantified severity values, it pairs the coordinates of all candidate points, such as Shenque (CV8) and Guanyuan (CV4), with their severity values ​​to form a list. Then, a quicksort algorithm is used to sort them from highest to lowest value, generating a priority-sorted set of acupoint treatment point coordinates. For example, the highest priority point corresponds to coordinates (120, 150) with a severity of 0.9. This helps TCM practitioners prioritize high-risk areas. In one implementation, this acupoint treatment point coordinate set is further integrated into the diagnostic system, for example, by combining it with the user interface to display the sorted acupoints, helping to develop personalized treatment plans, such as recommending moxibustion for high-priority points, thereby improving overall diagnostic efficiency.

[0129] Preferably, the far-infrared intelligent moxibustion collaborative system based on big data analysis provided in this embodiment includes a priority sequence generation unit, a sensor feedback data stream acquisition unit, a dynamic adjustment treatment parameter generation unit, and an optimized temperature value acquisition unit in the temperature value acquisition module 40. The priority sequence generation unit is used to acquire the coordinate set of acupoint treatment points and calculate the thermal correlation degree to generate a priority sequence of the acupoint treatment point coordinate set.

[0130] The heat correlation is converted into a normalized priority using the softmax function according to the following formula, which is then used to generate a priority sequence:

[0131] (16)

[0132] In formula (16), Indicates the first The priority probability of each acupoint treatment point Indicates the first Thermal correlation of points Indicates temperature parameter, Indicates all categories of Summation, The total number of acupoints is represented. The control logic of formula (16) is based on the thermal correlation of each treatment point. Through the process of "thermal correlation index amplification → global normalization → priority probability generation", the thermal correlation is converted into a normalized priority probability, so as to realize the reasonable quantification and sorting of the priority of treatment points. The core is to use the normalization characteristics of the softmax function to make the treatment points with high thermal correlation obtain higher priority probability, while ensuring that the sum of the priority probabilities of all points is 1, which is convenient for generating an ordered priority sequence in the future.

[0133] No. The thermal correlation degree of each point is obtained by the following formula:

[0134] (17)

[0135] In formula (17), Indicates the first Average thermal correlation of each acupoint treatment point Indicates the first Coordinate vectors of acupoint treatment points Indicates the first Coordinate vectors of acupoint treatment points This represents the total number of coordinates of acupoint treatment points. The thermal diffusion parameter is represented. The control logic of formula (17) is based on the spatial coordinates of the acupoint treatment point. Through the process of "excluding its own point → calculating the spatial correlation between points → aggregating and averaging the correlation values", the average thermal correlation between a single treatment point and all other treatment points is quantified. The core is to use the Gaussian function to reflect the spatial thermal correlation characteristic of "the closer the distance, the stronger the correlation". After averaging, the overall thermal correlation degree of the point is obtained, which provides the core correlation basis for the subsequent generation of priority sequence.

[0136] The process of obtaining the coordinate set of acupoint treatment points by the priority sequence generation unit first involves extracting coordinate data from an infrared thermal imaging device, such as the scan results of abnormal areas on the human back. Assuming the coordinate set includes multiple points like (120, 150) and (180, 200), these points correspond to the locations of acupoints in Traditional Chinese Medicine, such as the Feishu (BL13) or Xinshu (BL15) acupoints. Next, thermal correlation is calculated. Thermal correlation refers to the correlation index of heat distribution between coordinate points, quantified by comparing the temperature difference between adjacent points. For example, if the temperature of the Feishu acupoint is 37.5 degrees Celsius, and the temperature of a neighboring point is 36.0 degrees Celsius, the correlation can be calculated as 0.8 based on the difference ratio, reflecting the tightness of heat conduction. Subsequently, a priority sequence is generated, sorting the coordinate sets from high to low according to thermal correlation. For example, the Feishu acupoint, with the highest correlation, is processed first to ensure strong treatment targeting. This method can optimize the allocation of thermotherapy resources in business operations and avoid inefficient treatment.

[0137] The sensor feedback data stream acquisition unit is used to control the hyperthermia execution unit according to the priority sequence and collect thermal radiation signals to obtain the sensor feedback data stream.

[0138] The sensor feedback data stream is derived using the following formula:

[0139] (18)

[0140] In formula (18), This represents the sensor feedback data stream. Indicates the first The thermal radiation signal at any given moment. The number of time points included in the priority sequence is indicated. The control logic of formula (18) is guided by the priority sequence and integrates the real-time thermal radiation feedback signals in the process of hyperthermia in an orderly manner through the process of "synchronous hyperthermia and acquisition → multi-time signal acquisition → data flow vector construction". This provides continuous monitoring data for subsequent dynamic adjustment of treatment parameters. The core is to synchronously control treatment and acquisition according to the priority sequence to ensure that the feedback data corresponds accurately with the treatment process.

[0141] No. The thermal radiation signal at a given time is obtained using the following formula:

[0142] (19)

[0143] In formula (19), Indicates the first Distribution of thermal radiation intensity at any given time. Indicates the collection area. This represents the spatial acquisition process of thermal radiation signals. The control logic of formula (19) is based on the regional thermal radiation intensity distribution at a certain moment. Through the process of "determining the acquisition area → summarizing the spatial thermal radiation intensity → outputting the signal at a single moment", the spatial thermal radiation distribution within the acquisition area is integrated into a single signal representing the overall thermal state at that moment. The core is to use spatial integration to summarize the total amount of thermal radiation within the area, avoid interference from single-point intensity fluctuations, and accurately reflect the overall thermal radiation level of the acquisition area at that moment.

[0144] When controlling the hyperthermia execution unit, the sensor feedback data stream acquisition unit activates devices such as infrared lamps or laser hyperthermia devices according to a priority sequence. It first applies heat stimulation to the first point in the sequence, while simultaneously acquiring heat radiation signals through built-in sensors to obtain a real-time data stream. For example, an increase in signal intensity from an initial 50 units to 80 units indicates effective heat penetration. This acquisition process ensures continuous feedback, providing fundamental data for subsequent analysis.

[0145] The dynamic adjustment treatment parameter generation unit is used to fit the temperature gradient curve based on the sensor feedback data stream. If the rate of change of the temperature gradient curve exceeds the threshold, dynamic adjustment treatment parameters are generated.

[0146] The dynamic adjustment of treatment parameters is derived using the following formula:

[0147] (20)

[0148] In formula (20), This indicates the dynamically adjusted treatment parameters, such as thermotherapy power and energy output, which represent the final adjustment result. This indicates the original treatment parameters, or the initial parameter values ​​before adjustment. This indicates the adjustment ratio coefficient, which controls the magnitude of parameter adjustment and needs to be preset according to the clinical scenario or equipment characteristics. It represents the rate of change of the temperature gradient curve, the real-time monitoring rate of temperature change, and reflects the dynamic state of the thermotherapy process. Indicates the preset threshold. Indicates an indicator function, when Greater than The value is 1 if the temperature gradient change rate is exceeded, otherwise it is 0. The control logic of formula (20) is based on the threshold of the temperature gradient change rate as the trigger condition. Through the process of "change rate threshold judgment → adjustment amount calculation → parameter update", the treatment parameters are dynamically adjusted only when the temperature change exceeds the limit, so as to ensure the temperature stability and safety of the thermotherapy process. The core is to use the indicator function to limit the adjustment trigger condition and combine the degree of exceedance to quantify the adjustment range, so as to achieve "dynamic adjustment on demand".

[0149] The rate of change of the temperature gradient curve is derived using the following formula:

[0150] (twenty one)

[0151] In formula (21), Represents the temperature gradient curve. The current moment is indicated. The control logic of formula (21) is based on the fitted temperature gradient curve. Through the process of "curve differentiation → rate of change extraction", the rate of change of the temperature gradient with time is directly obtained, which provides the core dynamic change index for subsequent determination of whether to adjust the treatment parameters. The core is to use the time derivative to quantify the real-time change rate of the temperature gradient.

[0152] The temperature gradient curve is derived using the following formula:

[0153] (twenty two)

[0154] In formula (22), This indicates the number of data points in the sensor feedback data stream. Indicates the first Each sensor data value, Indicates the first The time points corresponding to each data point This represents the temperature gradient function to be fitted. This indicates that the temperature gradient function to be fitted is at the th... The time point corresponding to each data point The predicted value at the location. The control logic of formula (22) is based on the time-data pairs fed back by the sensor. Through the process of “constructing the residual sum of squares target → minimizing the overall error → determining the fitting function”, the least squares method is used to fit the temperature gradient curve that fits the sensor data. The core is to minimize the residual sum of squares between the data points and the fitting curve so that the curve is as close as possible to all the feedback data and accurately reflects the trend of temperature gradient change over time.

[0155] The dynamic adjustment unit for generating treatment parameters fits a temperature gradient curve based on sensor feedback data streams. This involves using fitting techniques such as least squares to connect data points into a smooth curve. For example, the smooth curve shows the temperature gradually increasing from 36.0 degrees Celsius to 38.0 degrees Celsius from 0 to 5 minutes after the start of treatment. If the rate of change exceeds a preset threshold, such as 0.5 degrees Celsius per minute, dynamic adjustments to the treatment parameters are generated, such as reducing power to prevent overheating. This adjustment mechanism enables precise control in practice, reducing the risk of burns and improving treatment efficacy.

[0156] The optimized temperature value acquisition unit is used to feed back dynamically adjusted treatment parameters to the thermotherapy execution unit for iterative approximation to obtain the optimized temperature value.

[0157] The optimal temperature value is obtained using the following formula:

[0158] (twenty three)

[0159] In formula (23), This represents the optimized temperature value; the updated temperature is the input for the next iteration. This represents the temperature value before optimization, the initial temperature of the current iteration, which comes from the output or initial setting of the previous round. This indicates that the step size coefficient is adjusted to control the range of temperature changes with treatment parameters. It needs to be preset according to the equipment response characteristics and clinical needs. The dynamically adjusted treatment parameters, such as the output of formula (20) and the energy output, are the driving force for temperature adjustment. The control logic of formula (23) is based on iterative temperature approximation optimization. By converting the dynamically adjusted treatment parameters into gradual temperature corrections, the ideal treatment temperature is approximated in an iterative manner, thereby achieving precise temperature control in the thermotherapy process. Specifically, based on the current temperature and the dynamically adjusted treatment parameters, the adjustment of treatment parameters is converted into gradual temperature corrections through the process of "adjustment amount calculation → temperature iterative update → cyclic approximation optimization". Finally, the ideal optimized temperature value is iteratively approximated. The core is to use the adjustment step size coefficient to control the temperature change range with the parameters, and to achieve precise temperature optimization through iteration.

[0160] The process by which the temperature acquisition unit dynamically adjusts treatment parameters and feeds them back to the thermotherapy execution unit for iterative approximation is similar to closed-loop control. For example, if the initial temperature is set to 37.0 degrees Celsius, it is adjusted to 37.2 degrees Celsius after feedback. Through multiple iterations, such as three rounds of approximation, an optimized temperature value, such as 37.5 degrees Celsius, is finally obtained, ensuring thermal balance at the treatment point. This iteration provides stable temperature management in practice, supporting the reliability of long-term thermotherapy applications.

[0161] Furthermore, the far-infrared intelligent moxibustion collaborative system based on big data analysis provided in this embodiment includes a moxibustion precision treatment distance parameter determination module 50 comprising an associated dataset establishment unit, a three-dimensional spatial calibration instruction generation unit, a real-time physical distance data acquisition unit, and a moxibustion precision treatment distance parameter output unit. The associated dataset establishment unit is used to receive optimized temperature values ​​and trigger the distance sensor to collect real-time physical distance data to establish an associated dataset of temperature and distance.

[0162] The correlation dataset between temperature and distance is derived using the following formula:

[0163] (twenty four)

[0164] In formula (24), This indicates the degree of matching between the temperature and distance datasets. This represents the total number of samples collected. Indicates the optimized temperature value and corresponding spacing The matching indicator function. The control logic of formula (24) is based on the collected temperature-distance sample pairs. Through the process of "sample matching judgment → matching quantity aggregation → matching degree calculation", the reliability of the association between temperature and distance is quantified, and samples that meet the preset association relationship are selected to establish a reliable temperature-distance association dataset. The core is to use the matching indicator function to distinguish between valid / invalid samples, and to reflect the association matching degree through the proportion, so as to ensure the reliability of the dataset.

[0165] The process of establishing a correlated dataset by receiving optimized temperature values ​​first involves obtaining values ​​from the preceding thermotherapy module. For example, in a treatment scenario targeting acupoints like Dazhui (GV14) in Traditional Chinese Medicine, the optimized temperature value is set to 38.5 degrees Celsius. The system automatically triggers the distance sensor to start data acquisition. The distance sensor monitors the physical distance between the treatment device and the patient's skin in real time using laser ranging principles. For instance, if a data stream with a distance of 5.0 cm is collected during treatment, this data is combined with the temperature intensity to establish a correlated dataset. Specifically, temperature intensity refers to the energy level of thermal radiation, and its correlation with spatial distance is constructed through a mapping relationship. For example, when the temperature is 38.5 degrees Celsius, a distance of 5.0 cm corresponds to an intensity value of 80 units, while increasing the distance to 6.0 cm reduces the intensity to 70 units, forming a dataset table for subsequent analysis. This process ensures the safety of thermotherapy in business operations, avoiding heat attenuation caused by improper distance.

[0166] The three-dimensional spatial calibration instruction generation unit is used to call the positioning system to obtain spatial coordinates and calculate the attitude offset vector based on the associated dataset. If the attitude offset vector is greater than a preset threshold, a three-dimensional spatial calibration instruction is generated.

[0167] The three-dimensional spatial calibration command is derived from the following formula:

[0168] (25)

[0169] In formula (25), This indicates the generated three-dimensional space calibration command. Represents the attitude offset vector. This indicates the preset threshold. This function generates a corresponding calibration command based on the current offset. The calibration command is triggered when the magnitude of the attitude offset vector is greater than the threshold; otherwise, no command is generated. The control logic of formula (25) uses the over-limit judgment of the attitude offset vector as the trigger condition. Through the process of "offset calculation → threshold comparison → command generation on demand", the calibration command is generated only when the positioning attitude deviates from the reasonable range, ensuring the accuracy of the three-dimensional spatial positioning of the moxibustion device. The core is to use the magnitude of the attitude offset vector to measure the degree of deviation, avoid meaningless frequent calibration, and achieve accurate spatial attitude calibration.

[0170] The magnitude of the attitude offset vector is obtained by the following formula:

[0171] (26)

[0172] In formula (26), This represents the magnitude of the attitude offset vector. , , These represent the offset vectors in three-dimensional space. , , The formula calculates the Euclidean norm of the attitude offset vector, i.e., the total offset, for comparison with a preset threshold. The control logic of formula (26) is based on the three-dimensional components of the attitude offset vector. Through the process of "component square calculation → sum of squares → square root norm calculation", the three-dimensional offset is integrated into a modulus representing the degree of total offset, providing a unified quantitative index for subsequent attitude offset threshold determination. The core is to calculate the Euclidean norm, which transforms the dispersed offset in the three-dimensional direction into a single total offset, making it easy to compare with the threshold.

[0173] The 3D spatial calibration command generation unit calls the positioning system based on the associated dataset. The positioning system uses GPS or indoor infrared positioning technology to obtain spatial coordinates, such as the current coordinates of the treatment device being (100, 150, 200) and the coordinates of the acupoints on the human body being (105, 155, 205). Then, it calculates the attitude offset vector, that is, the difference vector between the two coordinates, for example, a vector value of (5, 5, 5). If its magnitude is greater than a preset threshold, such as 3 units, a 3D spatial calibration command is generated. This 3D spatial calibration command includes rotation and translation parameters to adjust the device's attitude. In business applications, this calculation helps with real-time calibration, ensuring treatment accuracy, especially preventing deviation amplification when the patient makes slight movements.

[0174] The real-time physical distance data acquisition unit is used to drive the execution unit to complete the pose correction and acquire the corrected real-time physical distance data according to the three-dimensional space calibration command.

[0175] The corrected pose is obtained using the following formula:

[0176] (27)

[0177] In formula (27), This indicates the corrected pose. This represents a rotation matrix driven by a three-dimensional spatial calibration command. Indicates the initial pose. This represents the translation vector driven by the three-dimensional spatial calibration command. The control logic of formula (27) is based on the pose parameters and initial pose of the calibration command. Through the process of "analyzing calibration parameters → rotating to correct the attitude → translating to correct the position → combining to obtain the corrected pose", the three-dimensional spatial calibration command is transformed into the accurate pose correction of the device, providing a pose basis for obtaining accurate real-time physical distance data. The core is to correct the attitude angle through the rotation matrix and correct the spatial position through the translation vector to achieve accurate calibration of the device pose.

[0178] The corrected real-time physical spacing data is obtained using the following formula:

[0179] (28)

[0180] In formula (28), This indicates the corrected real-time physical spacing. Represents the three-dimensional coordinates of the first point. The three-dimensional coordinates of the second point are represented. The control logic of formula (28) is based on the three-dimensional coordinates of the device and the treatment point after correction. Through the process of "extracting coordinate components → calculating the distance difference of each axis → integrating the three-dimensional distance contribution → calculating the straight line distance", the real-time physical distance between the corrected moxibustion device and the acupoint treatment point is calculated. This provides direct spatial distance data for determining the accurate treatment distance parameters. The core is to calculate the Euclidean straight line distance between two points in three-dimensional space and accurately quantify the actual distance between the device and the treatment point.

[0181] The process by which the real-time physical distance data acquisition unit drives the execution unit to complete pose correction based on three-dimensional spatial calibration instructions is similar to robotic arm control. For example, after receiving the instruction, the execution unit adjusts the angle and position through servo motors, correcting the offset vector (5, 5, 5) to an approximate zero vector. Then, it re-acquires the corrected real-time physical distance data, such as adjusting it from the original 5.0 cm to a precise 4.5 cm. This correction mechanism improves stability in thermotherapy operations, reduces errors caused by human intervention, and provides a reliable foundation for subsequent fusion.

[0182] The Danzhi Precision Treatment Distance Parameter Output Unit is used to fuse optimized temperature values ​​and corrected real-time physical distance data, calculate and output the Danzhi precision treatment distance parameters.

[0183] The precise treatment distance parameter for moxibustion is derived using the following formula:

[0184] (29)

[0185] In formula (29), represents the precise moxibustion treatment distance parameter,[ represents the current temperature value after fusion,[ represents the exponential decay rate of the temperature deviation; This formula reflects the relationship that the greater the temperature deviates from the optimal value, the exponentially decreasing safety and effective treatment distance. The control logic of formula (29) takes the corrected physical distance and the degree of temperature deviation as the core, and dynamically adjusts the treatment distance through the process of "calculating the temperature deviation amount → constructing the exponential decay term → fusing the influence of distance and temperature": the more the temperature deviates from the optimized value, the more exponentially the treatment distance shrinks, and finally outputs the precise moxibustion treatment distance parameter that takes into account both safety and effectiveness. The core is to use the exponential decay term to correlate the temperature deviation and the distance adjustment, realizing the safety treatment logic of "the more the temperature deviates, the more conservative the distance".

[0186] The calculation of the precise moxibustion treatment distance parameter output unit for fusing the optimized temperature value and the corrected distance data involves a weighted average algorithm. For example, combining the temperature value of 38.5 degrees Celsius with the distance data of 4.5 centimeters, the precise moxibustion treatment distance parameter of 4.2 centimeters is output, and this precise moxibustion treatment distance parameter guides the treatment distance of moxibustion. In one embodiment, this output realizes personalized treatment in business, such as acupoint therapy for human body humidity, ensuring uniform penetration of heat and bringing better therapeutic effect recovery. In one embodiment, when extended to multi-device linkage, the associated data set can further integrate multi-sensor inputs, such as combining the dynamic changes of temperature and distance to form a closed-loop feedback, supporting the comprehensive treatment of complex acupoints such as Fengchi acupoint, thereby expanding the application scope in business and improving the overall hyperthermia efficiency.

[0187] Preferably, for the far-infrared intelligent moxibustion collaborative system based on big data analysis provided in this embodiment, the integrated operation execution module 60 includes a continuous collaborative treatment sequence generation unit, an actual motion trajectory model construction unit, and a correction instruction execution unit. Among them, the continuous collaborative treatment sequence generation unit is used to obtain the precise moxibustion treatment distance parameter, the duration control logic, and the vector data of the posture adaptation mechanism, and map the vector data of the posture adaptation mechanism to generate a continuous collaborative treatment sequence.[

[0188] The continuous collaborative treatment sequence is obtained through the following formula:[

[0189] (30)

[0190] In formula (30), represents the continuous collaborative treatment sequence,[ The initial vector representing the attitude adaptation mechanism is generated collaboratively by the system's acupoint coordinate mapping module and the initial attitude template library. The endpoint vector representing the posture adaptation mechanism is generated by the system's moxibustion treatment plan database and acupoint coordinate calibration module. Indicates the mapping interpolation parameters. The duration is calculated and generated in real time by the system's duration control logic module. If the preset total duration of this treatment segment is T, then... The duration control logic module will drive the treatment according to the requirements of the moxibustion treatment plan (such as uniform speed treatment or staged variable speed treatment). exist Gradual transition within the interval. The control logic of formula (30) is based on the initial / endpoint vector of the posture adaptation mechanism as the boundary and the normalized interpolation parameter as the time-driven mechanism. Through linear interpolation, the posture is smoothly connected to generate a continuous and coordinated treatment sequence that matches the duration and distance requirements of moxibustion treatment. The core is to use the gradual change of the normalized parameter to realize the continuous and controllable transition of the mechanism posture from the initial to the end point, and to ensure the stability of the moxibustion treatment action.

[0191] The process of obtaining precise treatment distance parameters for moxibustion using the continuous collaborative treatment sequence generation unit first extracts relevant values ​​from the core database of the treatment device. For example, in a scenario involving acupoints like Zusanli (ST36), the distance parameter is set to 3.5 cm. The system then synchronously invokes the duration control logic, which is based on the thermal sensitivity of the acupoint, for example, controlling the treatment duration to within 15 minutes to avoid overheating. Subsequently, vector data from the posture adaptation mechanism is collected through multi-axis sensors. For example, the current posture of the posture adaptation mechanism is represented as a vector (2, 3, 4). This vector data is integrated to provide the basis for subsequent mapping. Specifically, the duration control logic is a time-series-based algorithm that dynamically adjusts according to the acupoint type. For example, for Zusanli (ST36), if the initial duration is 15 minutes, the logic monitors the thermal feedback signal. If the feedback shows stable temperature, it extends to 18 minutes. This adjustment ensures the continuity of treatment.

[0192] When mapping the vector data of the posture adaptation mechanism to generate a continuous collaborative treatment sequence, a coordinate transformation method is used. First, the vector (2, 3, 4) is projected onto the treatment plane to form a sequence of points, such as a path sequence from the starting point to the ending point. These sequence of points are then connected into a continuous curve using an interpolation algorithm. In one embodiment, this mapping is used in practice for multi-acupoint linkage, such as combining Zusanli and Hegu acupoints, to generate a collaborative sequence that ensures a smooth transition when the device moves, avoiding treatment interruptions.

[0193] The actual motion trajectory model construction unit is used to analyze the continuous coordinated treatment sequence to obtain the underlying pulse control signal, drive the posture adaptation mechanism to operate based on the underlying pulse control signal, and construct the actual motion trajectory model.

[0194] The underlying pulse control signal is derived using the following formula:

[0195] (31)

[0196] In formula (31), This indicates the underlying pulse control signal. The decoding mapping function, representing the conversion from a continuous synergistic treatment sequence to a pulse signal, is a predefined hardware calibration function of the system. It is pre-calibrated based on the hardware characteristics of the actuators of the attitude adaptation mechanism. For the actuators such as the stepper motor and servo motor of the attitude adaptation mechanism, calibration experiments of "control parameters - physical actions" are carried out (such as testing the "correspondence between the number of pulses and the amount of displacement" and the "correspondence between the pulse duration and the servo motor angle"). The calibration results are encapsulated into mapping rules (including numerical conversion and unit matching logic) and written into the system control module to form the decoding mapping function. The predefined pulse sequence template represents a standard pulse parameter framework corresponding to different action types of the posture adaptation mechanism. It includes the basic pulse format (such as square wave) and configurable parameter fields (such as frequency, duty cycle, and duration). Different actions (such as "three-dimensional position movement" and "posture angle deflection") correspond to different templates. The control logic of formula (31) takes the continuous collaborative treatment sequence and the predefined pulse template as input. Through the process of "template loading → sequence decoding mapping → real-time pulse generation", the posture instructions of moxibustion treatment are converted into the underlying pulse signals that drive the operation of the posture adaptation mechanism. The core is to use the decoding function to associate the posture requirements with the hardware control parameters to ensure that the posture adaptation mechanism accurately executes the treatment actions.

[0197] The actual motion trajectory model is derived using the following formula:

[0198] (32)

[0199] In formula (32), Represents the actual motion trajectory model, at time... The three-dimensional position vector of the posture adaptation mechanism reflects the real-time motion path of the mechanism during moxibustion treatment. This indicates the initial position, which is the three-dimensional spatial coordinate when the posture adaptation mechanism begins to execute the continuous synergistic treatment sequence. It is the starting point and benchmark of the motion trajectory model, with dimensions and Consistency includes core parameters such as the initial horizontal position and vertical height of the posture adaptation mechanism relative to the target acupoint, which are obtained collaboratively by the system's acupoint positioning module and the mechanism calibration unit.

[0200] Represents the attitude-based velocity function, given by time... Institutional posture The determined instantaneous velocity vector is a dynamic mapping function of "posture-velocity," outputting vector parameters that include the magnitude of the velocity and the direction of motion. Its core characteristic is that different treatment postures correspond to different movement speeds (such as low-speed fine adjustment when "approaching the acupoint," and high-speed movement when "translating and changing position"), ensuring the stability and precision of moxibustion treatment movements. The data is obtained by generating the system attitude-velocity mapping table and the real-time attitude analysis unit. The current moment is indicated. The control logic of formula (32) is based on the initial position and the velocity associated with the attitude as the dynamic drive. Through the process of "initial reference positioning → attitude-velocity association → time integration to calculate displacement → trajectory model construction", the actual motion path of the attitude adaptation mechanism is restored. The core is to use the constraint of attitude on velocity and combine time integration to accumulate displacement to accurately reproduce the actual motion trajectory of the attitude adaptation mechanism.

[0201] The actual motion trajectory model construction unit analyzes the continuous coordinated treatment sequence to obtain the underlying pulse control signal, which involves signal decomposition technology. For example, the sequence curve is decomposed into pulse waveforms, each waveform corresponding to a control command. For instance, a pulse with a frequency of 50 Hz is used to drive the motor to rotate. This analysis process extracts parameters from the geometric features of the sequence to ensure the accuracy of the signal. In one embodiment, when the posture adaptation mechanism is driven to operate based on the underlying pulse control signal and the actual motion trajectory model is constructed, the signal is transmitted to the actuator motor. For example, the pulse signal drives the mechanism to move from vector (2, 3, 4) to (3, 4, 5), and the motion path is recorded to form a trajectory model. This model reflects the actual operating state by fitting a curve through sampling points.

[0202] The correction instruction execution unit is used to calculate the deviation matrix between the actual motion trajectory model and the continuous synergistic treatment sequence, generate a compensation vector based on the deviation matrix, and execute correction instructions to ensure treatment accuracy and adaptability.

[0203] The deviation matrix is ​​obtained by the following formula:

[0204] (33)

[0205] In formula (33), The deviation matrix is ​​represented. The control logic of formula (33) takes the actual motion trajectory and the target treatment sequence as inputs, and quantifies the degree of deviation between the actual action and the target instruction through the process of "vector alignment → dimensional difference calculation → deviation matrix construction", providing accurate deviation data for subsequent treatment correction. The core is to directly compare the actual and target state parameters to intuitively reflect the deviation of each treatment dimension.

[0206] The correction command is derived using the following formula:

[0207] (34)

[0208] In formula (34), Indicates a correction instruction. This represents the original command, which is the original control command without deviation calibration. It corresponds to the target action parameters of the continuous collaborative treatment sequence and is generated collaboratively by the continuous collaborative treatment sequence generation unit and the underlying signal decoding module. The compensation vector is represented. The control logic of formula (34) is based on the original instruction and the compensation vector is used as the basis for deviation correction. Through the process of "instruction-vector dimension alignment → vector superposition correction → output execution instruction", the deviation compensation amount is integrated into the original instruction to realize the calibration of the actual motion trajectory to the target treatment sequence. The core is to quickly make up for the deviation between the actual action and the target instruction through simple vector superposition, so as to ensure the accuracy of moxibustion treatment.

[0209] The compensation vector is obtained using the following formula:

[0210] (35)

[0211] In formula (35), The number of columns in the deviation matrix is ​​obtained through the system's matrix dimension statistics module. Indicates all The column vector is used to sum the multi-time deviation values ​​of each row of the deviation matrix, and is predefined and generated by the system's mathematical tool module. This formula calculates the average value of each row of the deviation matrix to generate the compensation vector. The control logic of formula (35) is based on the multi-time deviation data of the deviation matrix. Through the process of "dimension-time deviation aggregation → row average smoothing → compensation vector generation", a stable deviation compensation amount is obtained. The core is to take the average of the multi-time deviations of the same control dimension to avoid the random fluctuation of a single deviation and ensure the stability and rationality of the compensation for moxibustion treatment.

[0212] The process by which the correction instruction execution unit calculates the deviation matrix between the actual motion trajectory model and the continuous coordinated treatment sequence includes point-to-point comparisons. For example, if the model trajectory points are (3.2, 4.1, 5.0) and the sequence points are (3, 4, 5), the deviation is calculated as a difference matrix, such as [[0.2], [0.1], [0]]. This matrix quantifies the degree of offset. Specifically, when generating a compensation vector based on the deviation matrix and executing correction instructions to ensure treatment accuracy and adaptability, a vector such as (0.2, 0.1, 0) is extracted from the deviation matrix, and then a reverse compensation instruction is generated, such as adjusting the x-axis by -0.2. This closed-loop correction improves the reliability of the operation, especially maintaining treatment consistency when the patient's position changes.

[0213] The present invention also provides an intelligent moxibustion bed, including an intelligent bed body and a far-infrared intelligent moxibustion coordination system based on big data analysis installed on the intelligent bed body, which will not be described in detail here.

[0214] This embodiment provides a far-infrared intelligent moxibustion collaborative system and intelligent moxibustion bed based on big data analysis. Compared with existing technologies, it addresses the core business scenario problem of insufficient treatment precision caused by uneven distribution of physiological thermal energy in the human body, and the impact of posture deviation on treatment effects. It achieves precise treatment through multi-dimensional data fusion and dynamic calibration mechanisms. First, a uniform acupoint distribution map is generated through thermal radiation scanning. Combined with analysis algorithms and physiological model comparison, thermal anomaly areas are accurately marked and the coordinate set of treatment points is extracted. Second, based on priority ranking and sensor feedback, the temperature gradient curve is dynamically calculated to optimize treatment parameters. Finally, distance sensor data and posture calibration system are fused to determine the precise treatment distance, and duration control and vector data are integrated to form a continuous collaborative treatment sequence, ensuring treatment accuracy and adaptability. The overall technical effect of this embodiment is that through the integrated innovation of thermal energy distribution analysis, dynamic parameter adjustment, and posture calibration, it significantly improves the accuracy and personalized adaptability of acupoint treatment, providing an efficient solution for precise intervention of thermal energy anomalies in the human physiological system.

[0215] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention. Clearly, those skilled in the art can make various alterations and modifications to the invention without departing from its spirit and scope. Thus, if these modifications and modifications of the invention fall within the scope of the claims and their equivalents, the invention is also intended to include these modifications and modifications.

Claims

1. A far-infrared intelligent moxibustion collaborative system based on big data analysis, characterized in that, include: The homogenized acupoint distribution map generation module (10) is used to perform thermal radiation scanning on acupoints on the human body surface using a radiation detection device, collect physiological thermal data, and generate a homogenized acupoint distribution map through smoothing processing. The homogenized acupoint distribution map reflects the thermal energy state of the human physiological system. The abnormal area marking module (20) is used to compare the uniform acupoint distribution map with the preset physiological model template by applying the analysis algorithm to detect thermal abnormal deviation. When the thermal abnormal deviation exceeds the range, the abnormal area is marked. Priority sorting module (30) is used to analyze the thermal gradient changes in the abnormal area using a pattern matching algorithm, extract the coordinate set of acupoint treatment points and sort them according to priority, wherein the priority takes into account the severity of the abnormality; The optimized temperature value acquisition module (40) is used to acquire the priority sequence of the acupoint treatment point coordinate set, link with the sensor feedback of the thermotherapy execution unit, calculate the temperature gradient curve, and dynamically adjust the treatment parameters to obtain the optimized temperature value. The precise treatment distance parameter determination module (50) for moxibustion is used to integrate the optimized temperature value and distance sensor data, evaluate the influence of posture deviation, and determine the precise treatment distance parameter for moxibustion by calibrating the positioning system if the posture deviation causes a deviation. The integrated operation execution module (60) is used to integrate the vector data of the precise treatment distance parameters, duration control logic and posture adaptation mechanism of the moxibustion, form a continuous collaborative treatment sequence, and perform integrated operation to ensure treatment accuracy and adaptability.

2. The far-infrared intelligent moxibustion collaborative system based on big data analysis according to claim 1, characterized in that, The homogenized acupoint distribution map generation module (10) includes: The two-dimensional scanning matrix conversion unit is used to acquire the radiation intensity signal collected by the radiation detection device and convert the radiation intensity signal into a two-dimensional scanning matrix according to the acupoint coordinate mapping relationship. A continuous gradient distribution model generation unit is used to perform median filtering and bicubic interpolation calculations on the two-dimensional scanning matrix to generate a continuous gradient distribution model. The uniform acupoint distribution map generation unit is used to process the continuous gradient distribution model using pixel mapping technology to generate a uniform acupoint distribution map, wherein the uniform acupoint distribution map is enhanced in contrast by histogram equalization. The thermal energy state determination unit is used to extract the state characteristics of the homogenized acupoint distribution map and determine the thermal energy state of the human physiological system based on the state characteristics.

3. The far-infrared intelligent moxibustion collaborative system based on big data analysis according to claim 1, characterized in that, The abnormal region marking module (20) includes: The standard reference thermal field generation unit is used to obtain a homogenized acupoint distribution map and load a preset physiological model template to generate a standard reference thermal field with spatial alignment. The gray-scale difference matrix construction unit is used to perform differential calculations on the homogenized acupoint distribution map and the standard reference thermal field to construct a gray-scale difference matrix; The thermal anomaly deviation distribution set generation unit is used to weight the gray-level difference matrix by combining the regional sensitivity coefficient to generate the thermal anomaly deviation distribution set. An abnormal feature point cloud set extraction unit is used to extract an abnormal feature point cloud set if the thermal anomaly deviation distribution set exceeds a preset threshold. An abnormal region marking unit is used to process the abnormal feature point cloud set to extract edge contours and mark abnormal regions on the homogenized acupoint distribution map.

4. The far-infrared intelligent moxibustion collaborative system based on big data analysis according to claim 1, characterized in that, The priority sorting module (30) includes: The local thermal gradient vector field calculation unit is used to calculate the local thermal gradient vector field of the pixel grayscale matrix within the abnormal region. The local thermal gradient vector field reflects the directional change rate of the pixel. The candidate treatment point center coordinate positioning unit is used to perform normalized cross-correlation matching between the local thermal gradient vector field and the pre-set acupoint thermal feature template library to locate the center coordinates of the candidate treatment point. The quantitative anomaly severity value acquisition unit is used to acquire the quantitative anomaly severity value corresponding to the center coordinates of the candidate treatment point. The quantitative anomaly severity value represents the degree of deviation of the gradient magnitude from the preset standard physiological model. The acupoint treatment point coordinate set generation unit is used to sort the center coordinates of the candidate treatment points in descending order according to the quantified severity of the abnormality value, and generate a set of acupoint treatment point coordinates sorted by priority.

5. The far-infrared intelligent moxibustion collaborative system based on big data analysis according to claim 1, characterized in that, The optimized temperature value acquisition module (40) includes: The priority sequence generation unit is used to obtain the coordinate set of acupoint treatment points and calculate the thermal correlation degree to generate a priority sequence of the acupoint treatment point coordinate set. The sensor feedback data stream acquisition unit is used to control the hyperthermia execution unit and collect thermal radiation signals according to the priority sequence to obtain the sensor feedback data stream. The dynamic adjustment treatment parameter generation unit is used to fit a temperature gradient curve based on the sensor feedback data stream, and generate dynamic adjustment treatment parameters if the rate of change of the temperature gradient curve exceeds a threshold. The optimized temperature value acquisition unit is used to feed back the dynamically adjusted treatment parameters to the thermotherapy execution unit for iterative approximation to obtain the optimized temperature value.

6. The far-infrared intelligent moxibustion collaborative system based on big data analysis according to claim 1, characterized in that, The precise treatment distance parameter determination module (50) for moxibustion includes: The associated dataset establishment unit is used to receive optimized temperature values ​​and trigger the distance sensor to collect real-time physical distance data to establish an associated dataset of temperature and distance. The three-dimensional space calibration instruction generation unit is used to call the positioning system to obtain spatial coordinates and calculate the attitude offset vector based on the associated dataset. If the attitude offset vector is greater than a preset threshold, a three-dimensional space calibration instruction is generated. The real-time physical distance data acquisition unit is used to drive the execution unit to complete the pose correction according to the three-dimensional space calibration command and acquire the corrected real-time physical distance data. The precise treatment distance parameter output unit for moxibustion is used to integrate the optimized temperature value and the corrected real-time physical distance data, calculate and output the precise treatment distance parameter for moxibustion.

7. The far-infrared intelligent moxibustion collaborative system based on big data analysis according to claim 1, characterized in that, The integrated operation execution module (60) includes: The continuous collaborative treatment sequence generation unit is used to acquire the precise treatment distance parameters, duration control logic, and vector data of the posture adaptation mechanism of Danzhi, and to map the vector data of the posture adaptation mechanism to generate a continuous collaborative treatment sequence. The continuous synergistic therapy sequence is derived using the following formula: ； in, Indicates a continuous synergistic treatment sequence. This represents the initial vector of the attitude adaptation mechanism. This represents the endpoint vector of the attitude adaptation mechanism. Indicates the mapping interpolation parameters; The actual motion trajectory model construction unit is used to analyze the continuous collaborative treatment sequence to obtain the underlying pulse control signal, drive the posture adaptation mechanism to operate based on the underlying pulse control signal, and construct the actual motion trajectory model. The correction instruction execution unit is used to calculate the deviation matrix between the actual motion trajectory model and the continuous synergistic treatment sequence, generate a compensation vector based on the deviation matrix, and execute correction instructions to ensure treatment accuracy and adaptability.

8. The far-infrared intelligent moxibustion collaborative system based on big data analysis according to claim 7, characterized in that, In the actual motion trajectory model construction unit, the underlying pulse control signal is obtained through the following formula: ； in, This indicates the underlying pulse control signal. This represents the decoding mapping function from a continuous synergistic therapy sequence to a pulse signal. This represents a predefined pulse sequence template; The actual motion trajectory model is derived using the following formula: ； in, This represents the actual motion trajectory model. Indicates the initial position. Represents the attitude-based velocity function. Indicates the current moment. Indicates time posture, Indicates time The rate of change of attitude.

9. The far-infrared intelligent moxibustion collaborative system based on big data analysis according to claim 8, characterized in that, In the correction instruction execution unit, the deviation matrix is ​​obtained by the following formula: ； in, Represents the deviation matrix; The correction command is derived using the following formula: ； in, Indicates a correction instruction. Indicates the original instruction. Represents the compensation vector; The compensation vector is obtained using the following formula: ； in, Indicates the number of columns in the deviation matrix. Indicates all Column vector.

10. An intelligent moxibustion bed, characterized in that, The invention includes a smart bed and a far-infrared smart moxibustion system based on big data analysis, as described in any one of claims 1 to 9, mounted on the smart bed.

Images