Analysis Method and System for the Correlation between Spinal-Pelvic Sagittal Morphological Parameters and Cardiopulmonary Exercise Endurance Parameters Based on Artificial Intelligence

Abstract

The present invention belongs to the technical field of medical image processing, and specifically relates to a method and system for analyzing the correlation between spinal-pelvic sagittal morphological parameters and cardiopulmonary exercise endurance parameters based on artificial intelligence. Specifically, by collecting spinal-pelvic sagittal morphological parameters and cardiopulmonary exercise endurance parameters, classifying the collected spinal-pelvic sagittal morphological parameters, using a neural network algorithm to extract relevant information features, and judging and outputting the correlation degree between the spinal-pelvic sagittal morphological parameters and the cardiopulmonary exercise endurance parameters through a correlation analysis model. The invention can objectively evaluate the correlation between the spinal-pelvic sagittal morphology and the cardiopulmonary exercise endurance, and provide a scientific basis for the individual's sports health management. By eliminating the interference of abnormal data, the accuracy and reliability of data analysis are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of medical image processing, and particularly relates to a method and system for analyzing the correlation between spinal-pelvic sagittal morphological parameters and cardiopulmonary exercise endurance parameters based on artificial intelligence. Background Art

[0002] With the rapid development of artificial intelligence technology, the acquisition and processing of relevant data have gradually become an important direction of research and application. On the basis of integrating multi-field data resources, it has become a trend to make full use of artificial intelligence technology to deeply analyze and solve complex problems.

[0003] In these efforts, the correlation analysis of spinal-pelvic sagittal morphological parameters and cardiopulmonary exercise endurance parameters is a field that has received much attention. On the one hand, spinal-pelvic sagittal morphological parameters directly reflect the physiological state and lesions of an individual's spine and pelvis; on the other hand, cardiopulmonary exercise endurance parameters are key indicators for evaluating cardiopulmonary function and are involved in multiple fields such as sports medicine, competitive sports training, and the prevention and treatment of cardiopulmonary diseases. In clinical diagnosis and sports medicine, it is often necessary to comprehensively evaluate an individual's motor function status and health condition, and abnormal spinal-pelvic morphology may reflect an individual's motor dysfunction or health risks.

[0004] Through correlation analysis, it is possible to understand whether abnormal spinal-pelvic morphology affects cardiopulmonary function, and thus provide an important reference for clinical diagnosis. At the same time, it can also provide data support for personalized exercise rehabilitation training programs. Especially for patients with spinal-pelvic diseases or abnormal cardiopulmonary function, the results of correlation analysis can guide medical staff to formulate corresponding exercise rehabilitation plans according to the different states of individuals, so as to achieve targeted treatment and training.

[0005] The present invention focuses on using artificial intelligence technology to analyze the correlation between an individual's spinal-pelvic sagittal morphological parameters and cardiopulmonary exercise endurance parameters, and thus provide data support for sports medicine and health management. Summary of the Invention

[0006] The object of the present invention is to provide a method and system for analyzing the correlation between spinal-pelvic sagittal morphological parameters and cardiopulmonary exercise endurance parameters based on artificial intelligence. Through data acquisition and processing, it can more accurately determine the significant correlation between spinal-pelvic sagittal morphological parameters and cardiopulmonary exercise endurance parameters, and provide a scientific basis for clinical medicine and sports science.

[0007] The technical solution adopted by the present invention is specifically as follows: A method for analyzing the correlation between spinal-pelvic sagittal morphological parameters and cardiopulmonary exercise endurance parameters based on artificial intelligence includes:

[0008] S1. Obtain spinal-pelvic sagittal morphological parameters and cardiopulmonary exercise endurance parameters;

[0009] S2. Preprocess the spinal-pelvic sagittal morphological parameters, where the preprocessing includes classification and denoising operations;

[0010] S3. Extract features based on an associated neural network: construct an associated neural network, and input the preprocessed spinal-pelvic sagittal morphological parameters and cardiopulmonary exercise endurance parameters into the associated neural network for feature extraction, and output the first feature information of the spinal-pelvic sagittal morphological parameters and the second feature information of the cardiopulmonary exercise endurance parameters;

[0011] S4. Output the correlation degree based on a preset correlation analysis model and feature information: obtain a preset correlation analysis model, and input the first feature information and the second feature information into the correlation analysis model to output the correlation degree between the spinal-pelvic sagittal morphological parameters and the cardiopulmonary exercise endurance parameters;

[0012] S5. Obtain a preset evaluation threshold and compare the correlation degree with the evaluation threshold; the comparison criteria include:

[0013] T1. If the correlation degree is greater than the evaluation threshold, it indicates that there is a significant correlation between the spinal-pelvic sagittal morphological parameters and the cardiopulmonary exercise endurance parameters;

[0014] T2. If the correlation degree is less than or equal to the evaluation threshold, it indicates that the correlation between the spinal-pelvic sagittal morphological parameters and the cardiopulmonary exercise endurance parameters is not significant.

[0015] In a preferred solution, the step of preprocessing the spinal-pelvic sagittal morphological parameters, where the preprocessing includes classification and denoising operations, includes:

[0016] Classify the spinal-pelvic sagittal morphological parameters into training sample data and target analysis data according to their sources, where the target analysis data is used as the analysis basis for the correlation between the subsequent spinal-pelvic sagittal morphological parameters and the cardiopulmonary exercise endurance parameters;

[0017] Determine the standard morphological parameters according to the training sample data, and then remove the abnormal data in the target analysis data based on the standard morphological parameters to obtain the normal data in the target analysis data;

[0018] Obtain a noise removal subset, and perform denoising processing on the normal data in the noise removal subset to obtain noise removal data;

[0019] Obtain multiple noise removal data within the monitoring period, arrange them in chronological order to obtain a sequence to be evaluated, then update the multiple noise removal data according to the sequence to be evaluated to obtain a pre-parameter, and use the pre-parameter as the analysis basis for the correlation between the spinal-pelvic sagittal plane morphological parameter and the cardiopulmonary exercise endurance parameter.

[0020] In a preferred embodiment, the step of performing denoising processing on the normal data in the noise removal subset to obtain noise removal data includes:

[0021] Obtain the noise removal subset;

[0022] Determine a plurality of first adjacent parameters according to the normal data in the noise removal subset;

[0023] Determine the corresponding pre-deviation amount according to the plurality of first adjacent parameters;

[0024] Obtain a denoising function, and input the pre-deviation amount and each normal data into the denoising function to obtain noise removal data.

[0025] In a preferred embodiment, the step of arranging multiple noise removal data within the monitoring period in chronological order to obtain a sequence to be evaluated, and then updating the multiple noise removal data according to the sequence to be evaluated to obtain a pre-parameter includes:

[0026] Determine the sequence to be evaluated according to the noise removal data;

[0027] Determine a plurality of second adjacent parameters according to the sequence to be evaluated;

[0028] Determine the corresponding post-deviation amount according to the plurality of second adjacent parameters;

[0029] Obtain an update function, and input the post-deviation amount and each noise removal data into the update function to obtain a plurality of pre-parameters.

[0030] In a preferred embodiment, the step of constructing a correlation neural network, inputting the preprocessed spinal-pelvic sagittal plane morphological parameter and cardiopulmonary exercise endurance parameter into the correlation neural network for feature extraction, and outputting the first feature information of the spinal-pelvic sagittal plane morphological parameter and the second feature information of the cardiopulmonary exercise endurance parameter includes:

[0031] Construct the correlation neural network, and use the preprocessed spinal-pelvic sagittal plane morphological parameter and cardiopulmonary exercise endurance parameter as inputs and transfer them to the input layer of the neural network;

[0032] The neural network is provided with multiple hidden layers, and transmits the spinal-pelvic sagittal morphological parameters and cardiopulmonary exercise endurance parameters in the input layer to the hidden layers, where non-linear transformation and feature extraction are performed;

[0033] Fuse the features output by multiple hidden layers and transmit them to the output layer to obtain the first feature information of the spinal-pelvic sagittal morphological parameters and the second feature information of the cardiopulmonary exercise endurance parameters.

[0034] In a preferred embodiment, the step of obtaining a preset correlation analysis model and inputting the first feature information and the second feature information into the correlation analysis model to output the correlation degree between the spinal-pelvic sagittal morphological parameters and the cardiopulmonary exercise endurance parameters includes:

[0035] Obtain a preset correlation analysis model, and input the first feature information and the second feature information into the correlation analysis model to obtain a first feature score corresponding to the first feature information and a second feature score corresponding to the second feature information;

[0036] Obtain an evaluation function, and input the first feature score and the second feature score into the evaluation function to obtain the correlation degree.

[0037] The present invention also provides a correlation analysis system for spinal-pelvic sagittal morphological parameters and cardiopulmonary exercise endurance parameters based on artificial intelligence, including:

[0038] A data acquisition module for acquiring spinal-pelvic sagittal morphological parameters and cardiopulmonary exercise endurance parameters;

[0039] A preprocessing module for preprocessing the spinal-pelvic sagittal morphological parameters, where the preprocessing includes classification and denoising operations;

[0040] A feature extraction module for constructing a correlation neural network, inputting the preprocessed spinal-pelvic sagittal morphological parameters and cardiopulmonary exercise endurance parameters into the correlation neural network for feature extraction, and outputting the first feature information of the spinal-pelvic sagittal morphological parameters and the second feature information of the cardiopulmonary exercise endurance parameters;

[0041] A correlation analysis module for obtaining a preset correlation analysis model and inputting the first feature information and the second feature information into the correlation analysis model to output the correlation degree between the spinal-pelvic sagittal morphological parameters and the cardiopulmonary exercise endurance parameters;

[0042] An evaluation module for obtaining a preset evaluation threshold and comparing the correlation degree with the evaluation threshold;

[0043] If the correlation degree is greater than the evaluation threshold, it indicates that there is a significant correlation between the spinal-pelvic sagittal morphological parameters and the cardiopulmonary exercise endurance parameters;

[0044] If the correlation degree is less than or equal to the evaluation threshold, it indicates that the correlation between the spinal-pelvic sagittal morphological parameters and the cardiopulmonary exercise endurance parameters is not significant.

[0045] The technical effects achieved by the present invention are as follows:

[0046] By performing a correlation analysis on the spinal-pelvic sagittal morphological parameters and the cardiopulmonary exercise endurance parameters, the present invention can provide a more comprehensive assessment of an individual's health status, improving the accuracy and reliability of diagnosis. At the same time, through the precise collection and effective denoising of data, the present invention reduces the influence of external interference on the analysis results, making the analysis results more credible. Through the correlation analysis, the individual's exercise and health response data can be accurately interpreted to help detect potential problems in a timely manner, such as abnormal functions of the spine and heart and lungs, providing a non-invasive and efficient means for individual health management. By using AI technology to improve the calculation efficiency, a large amount of data can be processed in a short time to generate personalized health reports and management plans. In addition, through the application of intelligent algorithms, the individual's health data can be continuously monitored and dynamically adjusted according to the change trend, thereby providing continuous technical support for individual health management. At the same time, in the field of sports medicine, it can provide a scientific basis for formulating personalized rehabilitation plans for athletes or patients, reducing blindness and risks. In clinical medicine, it can assist doctors in the early diagnosis and treatment planning of diseases, improving the level of medical services. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 It is a schematic flow chart of the analysis method of the present invention;

[0048] Figure 2 It is a schematic structural diagram of the analysis system of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0049] To more clearly illustrate the technical solution of the present invention, the following will detail the solution of the present invention in combination with embodiments. This embodiment is a detailed example based on the basic solution of the present invention, aiming to provide a supplementary description of the basic solution of the present invention.

[0050] See Figure 1 As shown, this embodiment provides a method for analyzing the correlation between spinal-pelvic sagittal morphological parameters and cardiopulmonary exercise endurance parameters based on artificial intelligence, including:

[0051] Obtain spinal-pelvic sagittal morphological parameters and cardiopulmonary exercise endurance parameters;

[0052] Obtain spinal-pelvic sagittal morphological parameters and simultaneously obtain cardiopulmonary exercise endurance parameters. The spinal-pelvic sagittal morphological parameters include, but are not limited to, pelvic tilt angle, spinal curvature, etc. The cardiopulmonary exercise endurance parameters include, but are not limited to, VO2max, heart rate recovery rate, maximum oxygen uptake, exercise time, exercise speed, exercise distance;

[0053] Preprocess the spinal-pelvic sagittal morphological parameters. The preprocessing includes classification and denoising operations;

[0054] Classify the collected spinal-pelvic sagittal morphological parameters (such as according to different anatomical sites), and perform denoising processing to eliminate possible outliers and noise during the collection process (such as using filters or statistical methods) to ensure the accuracy and reliability of the data;

[0055] Extract features based on an associated neural network: Construct an associated neural network, and input the preprocessed spinal-pelvic sagittal morphological parameters and cardiopulmonary exercise endurance parameters into the associated neural network for feature extraction, and output the first feature information of the spinal-pelvic sagittal morphological parameters and the second feature information of the cardiopulmonary exercise endurance parameters;

[0056] Construct an associated neural network model based on the spinal-pelvic sagittal morphological parameters and cardiopulmonary exercise endurance parameters, input the preprocessed parameters into this network, and perform feature extraction work to output the first feature of the spinal-pelvic sagittal morphological parameters and the second feature of the cardiopulmonary exercise endurance parameters;

[0057] Output the correlation degree based on a preset correlation analysis model and feature information: Obtain the preset correlation analysis model, and input the first feature information and the second feature information into the correlation analysis model to output the correlation degree between the spinal-pelvic sagittal morphological parameters and the cardiopulmonary exercise endurance parameters;

[0058] Establish a correlation analysis model, input the first feature and the second feature information into the correlation analysis model, so as to calculate the correlation degree between the spinal-pelvic sagittal morphological parameters and the cardiopulmonary exercise endurance parameters, providing data support for subsequent evaluation and optimization;

[0059] Obtain a preset evaluation threshold, and compare the correlation degree with the evaluation threshold;

[0060] Obtain a preset evaluation threshold, and compare and analyze the correlation degree according to the preset evaluation threshold (such as setting a certain correlation degree as a screening criterion);

[0061] If the correlation degree is greater than the evaluation threshold, it indicates that there is a significant correlation between the spinal-pelvic sagittal morphological parameters and the cardiopulmonary exercise endurance parameters;

[0062] If the correlation degree is less than or equal to the evaluation threshold, it indicates that the correlation between the spinal-pelvic sagittal plane morphological parameters and the cardiopulmonary exercise endurance parameters is not significant;

[0063] If the correlation degree is higher than the evaluation threshold, it can be considered that there is a significant correlation between the spinal-pelvic sagittal plane morphological parameters and the cardiopulmonary exercise endurance parameters. On the contrary, it indicates that the correlation between the two is not significant, and it may be necessary to further optimize the model or data processing.

[0064] In a preferred embodiment, preprocessing the spinal-pelvic sagittal plane morphological parameters, the preprocessing includes steps of classification and denoising, including:

[0065] Classify the spinal-pelvic sagittal plane morphological parameters into training sample data and target analysis data according to their sources. Among them, the target analysis data is used as the analysis basis for the correlation between the subsequent spinal-pelvic sagittal plane morphological parameters and the cardiopulmonary exercise endurance parameters;

[0066] Classify the spinal-pelvic sagittal plane morphological parameters into training sample data and target analysis data according to their sources. The training sample data is used to construct and calibrate the neural network model, and the target analysis data is used as the analysis basis for the correlation between the subsequent spinal-pelvic sagittal plane morphological parameters and the cardiopulmonary exercise endurance parameters;

[0067] Determine the standard morphological parameters according to the training sample data, and then remove the abnormal data in the target analysis data based on the standard morphological parameters to obtain the normal data in the target analysis data;

[0068] Determine the standard morphological parameters according to the training sample data. The standard parameters can be based on the statistical characteristics (such as mean, standard deviation, etc.) of the training sample data. Remove the outliers or data with large deviations in the target analysis data according to the standard morphological parameters to ensure the data quality and obtain the normal data set in the target analysis data;

[0069] Obtain the noise removal subset, and perform denoising processing on the normal data in the noise removal subset to obtain the noise removal data;

[0070] Obtain the preset noise removal subset, and perform further denoising processing (such as using filtering algorithms or transformation techniques) on the normal data in it. The noise removal work is to eliminate the random noise in the data and improve the accuracy of the data to obtain the data set after noise removal;

[0071] Obtain multiple noise-removed data within the monitoring period, arrange them in chronological order to obtain a sequence to be evaluated, then update the multiple noise-removed data based on the sequence to be evaluated to obtain a preparameter, and use the preparameter as the analysis basis for the correlation between the spinal-pelvic sagittal morphology parameter and the cardiopulmonary exercise endurance parameter.

[0072] By performing a time sorting on multiple noise-removed data within a preset monitoring period to form a sequence to be evaluated, and performing a merge-update operation on the multiple noise-removed data based on the sequence to be evaluated, it can subsequently be used as the analysis basis for the correlation between the spinal-pelvic sagittal morphology parameter and the cardiopulmonary exercise endurance parameter.

[0073] In a preferred embodiment, the step of determining the standard morphology parameter according to the training sample data and then removing the abnormal data in the target analysis data based on the standard morphology parameter to obtain the normal data in the target analysis data includes:

[0074] Determine multiple sample morphology parameters according to the training sample data;

[0075] Obtain the training sample data, construct conditions consistent with the target analysis data (that is, ensure that the conditions during the collection of the training sample data and the target analysis data are the same), so as to more accurately reflect the actual level of the target analysis data. Determine multiple sample morphology parameters according to the training sample data (for example: determine the pelvic tilt angle, pelvic rotation degree, and spinal curvature of the spinal-pelvic sagittal plane according to the training sample data). Under the same standard, the sample morphology parameters can be used as the reference parameters for subsequent measurement of the target analysis data;

[0076] Determine the sample fluctuation interval based on the multiple sample morphology parameters;

[0077] Determine the fluctuation amplitude of each sample morphology parameter according to the multiple sample morphology parameters. The determination method of the fluctuation amplitude, for example: by setting the fluctuation range of each morphology parameter. In this embodiment, a calculation method is used to determine the deviation, that is, to statistically calculate the median and its standard deviation of each sample morphology parameter data. The establishment of the sample fluctuation interval is to identify the standard morphology parameter, so as to be able to screen the target analysis data subsequently;

[0078] Obtain a sample evaluation function, and input the sample fluctuation interval into the sample evaluation function for calculation to obtain multiple sample evaluation values;

[0079] Obtain a preset sample evaluation function, input the multiple sample fluctuation intervals into the sample evaluation function for calculation, and obtain multiple sample evaluation values. The newly added calculation subsets need to meet the limitations of the sample fluctuation intervals. The number of the calculation subsets can be controlled according to the set time window. The output of the sample evaluation values can be used to judge the accuracy and reliability of the model, detect the performance of sensors or measurement devices, etc.;

[0080] Arrange the multiple sample evaluation values in descending order, and use the sample form parameter corresponding to the sample evaluation value with the highest ranking as the standard form parameter;

[0081] Arrange the multiple sample evaluation values in descending order, determine the longest time window, that is, the sample subset with the largest number of samples as the standard form parameter. The standard form parameter will be used as the benchmark calculation condition for subsequent target analysis data;

[0082] Obtain a screening interval, and screen the target analysis data under the standard form parameter according to the screening interval to obtain the normal data in the target analysis data.

[0083] Obtain a preset screening interval, screen the target analysis data under the standard form parameter condition to eliminate the abnormal data in the target analysis data, and obtain the normal data in the complete and effective target analysis data.

[0084] In a preferred embodiment, the step of determining the sample fluctuation interval according to the multiple sample form parameters includes:

[0085] Obtain the multiple sample form parameters;

[0086] Obtain a preset sample calculation duration, determine the calculation period of each sample form parameter according to the sample calculation duration, and then perform slicing processing based on the calculation period of each sample form parameter, so as to be able to divide the sample form parameter into multiple calculation subsets. It means that according to the calculation period of each sample form parameter, within how long the sample form parameter is divided into multiple sample form subsets. An advanced algorithm of International Data Corporation (IDC) needs to continuously re-evaluate itself. Today, understanding the needs and preferences of customers will change quantitatively, and the user experience may also change as the needs and preferences of customers change. It means that according to the calculation period of the sample form parameter, the calculation period is divided into multiple sample form subsets with equal duration (i.e., the calculation duration).

[0087] Determine the corresponding edge parameters according to the value length of each sample form parameter. Among them, the sample form parameter is the same as the edge parameter;

[0088] Obtain the length of each sample morphological parameter and record it as an edge parameter. The edge parameter is a core parameter used to determine the fluctuation value of the sample morphological parameter;

[0089] Obtain an edge measurement function, and input the edge measurement function and the edge parameter corresponding to the sample morphological parameter into the edge measurement function to obtain a sample fluctuation interval. Among them, the number of sample fluctuation intervals is the same as the number of sample morphological parameters.

[0090] Obtain a preset edge measurement function, and input the sample morphological parameter and the corresponding edge parameter into the edge measurement function together to measure the upper and lower fluctuation edges of the sample morphological parameter (for example: the fluctuation range of the sample morphological parameter can be calibrated up and down). The number of sample fluctuation intervals is the same as the number of input sample morphological parameters, that is, each sample morphological parameter can determine a sample fluctuation interval.

[0091] In a preferred embodiment, the step of performing denoising processing on the normal data in the noise removal subset to obtain noise removal data includes:

[0092] Obtain the noise removal subset;

[0093] Obtain a preset denoising operator and perform arithmetic processing in combination with the normal data in the target analysis data. The noise removal subset refers to the data obtained by screening the target normal data according to preset screening conditions during the monitoring period, and this part of the data can use the noise removal algorithm to remove random noise or outliers in the data;

[0094] Determine a plurality of first adjacent parameters according to the normal data in the noise removal subset;

[0095] Determining a plurality of first adjacent parameters according to the normal data in the noise removal subset can be achieved by identifying the similarity between adjacent samples in the noise removal data or by performing difference calculations to identify the similarity between the data to be evaluated and the abnormal data, providing basic data for subsequent dynamic sample updates;

[0096] Determine the corresponding pre-set deviation amount according to the plurality of first adjacent parameters;

[0097] Determine the corresponding pre-set deviation according to a plurality of first adjacent parameters, that is, before sample update, calculate the deviation value between adjacent samples as a reference value for subsequent updates, avoid the increase of similar or duplicate data, ensure the diversity and richness of the data set, and facilitate the analysis of different scenarios and situations.

[0098] Obtain a denoising function, and input the pre-set deviation amount and each normal data into the denoising function to obtain noise removal data.

[0099] Remove noise through the pre - deviation amount, normal data, and denoising function. At the same time, remove noise or inaccurate data in the positive data through the denoising algorithm to obtain relatively accurate and cleaned noise - removed data.

[0100] In a preferred embodiment, the step of obtaining a sequence to be evaluated according to multiple noise - removed data within a monitoring period, arranging them in chronological order, and then updating the multiple noise - removed data according to the sequence to be evaluated to obtain the pre - set parameters includes:

[0101] Determine the sequence to be evaluated according to the noise - removed data;

[0102] Obtain a preset condition to be evaluated, and determine the sequence to be evaluated according to the noise - removed data (for example, the amount of data in the sequence to be evaluated can be determined by setting a certain time - window screening condition). This sequence records the usage of the noise - removed data within the time from the data to be evaluated to the next change direction. In practical applications, the sequence to be evaluated can be divided into three categories: normal sequence to be evaluated, abnormal sequence to be evaluated, and noise sequence to be evaluated;

[0103] Determine multiple second - adjacent parameters according to the sequence to be evaluated;

[0104] Determine multiple second - neighbor parameters according to the sequence to be evaluated. Specifically, it refers to the distance between two adjacent samples in the sequence to be evaluated, aiming to select data with the closest adjacent parameters as much as possible for constructing the next matching data to improve the accuracy and effect of the update.

[0105] Determine the corresponding post - deviation amount according to the multiple second - adjacent parameters;

[0106] Determine the corresponding post - deviation according to multiple second - adjacent parameters, that is, determine the deviation value between the current sample data and the target analysis database according to the relationship of multiple second - adjacent parameters, enrich the diversity of the target analysis data, and be able to dynamically adjust the analysis samples;

[0107] Obtain an update function, and input the post - deviation amount and each noise - removed data into the update function to obtain multiple pre - set parameters.

[0108] Obtain an update function, and input the post - deviation amount in each noise - removed data into the update function to adjust the weight of the data in the sample database and determine the update direction of the sequence to be evaluated. Finally, we obtain multiple preliminary parameters. Based on these parameters, we can further find the sample with the highest similarity in the target analysis database for dynamic database analysis and update.

[0109] In a preferred embodiment, the step of constructing an associated neural network, inputting the preprocessed spinal-pelvic sagittal morphological parameters and cardiopulmonary exercise endurance parameters into the associated neural network for feature extraction, and outputting the first feature information of the spinal-pelvic sagittal morphological parameters and the second feature information of the cardiopulmonary exercise endurance parameters includes:

[0110] Construct the associated neural network, and use the preprocessed spinal-pelvic sagittal morphological parameters and cardiopulmonary exercise endurance parameters as inputs and transfer them to the input layer of the neural network;

[0111] Construct an associated neural network, merge the preprocessed spinal-pelvic sagittal morphological parameters and cardiopulmonary endurance parameters into one input, and transmit it to the input layer of the neural network to complete the construction and training of the neural network model;

[0112] The neural network has multiple hidden layers, and the spinal-pelvic sagittal morphological parameters and cardiopulmonary exercise endurance parameters in the input layer are transferred to the hidden layer, and non-linear transformation and feature extraction are performed in the hidden layer;

[0113] The neural network has multiple hidden layers, and the preprocessed spinal-pelvic sagittal morphological parameters and cardiopulmonary exercise endurance parameters are transferred to the hidden layer, and non-linear transformation and feature extraction are performed here, that is: the original data is converted into high-dimensional feature vectors through the processing of multiple hidden layers to provide support for subsequent classification and regression tasks;

[0114] Fuse the features output by multiple hidden layers and transfer them to the output layer to obtain the first feature information of the spinal-pelvic sagittal morphological parameters and the second feature information of the cardiopulmonary exercise endurance parameters;

[0115] Merge and fuse the feature information extracted by multiple hidden layers and transfer it to the output layer to obtain the first feature information of the spinal-pelvic sagittal morphological parameters and the second feature information of the cardiopulmonary exercise endurance parameters. Among them, the first feature information is used as the evaluation basis for the motor function state, and the second feature information is used as the cardiopulmonary function evaluation criterion.

[0116] In a preferred embodiment, the step of obtaining a preset correlation analysis model, inputting the first feature information and the second feature information into the correlation analysis model, and outputting the correlation degree between the spinal-pelvic sagittal morphological parameters and the cardiopulmonary exercise endurance parameters includes:

[0117] Obtain a preset correlation analysis model, input the first feature information and the second feature information into the correlation analysis model, and obtain a first feature score corresponding to the first feature information and a second feature score corresponding to the second feature information;

[0118] Obtain a preset correlation analysis model, and input the first feature information and the second feature information into the correlation analysis model to obtain a first feature score corresponding to the first feature information and a second feature score corresponding to the second feature information, objectively quantifying the individual's exercise parameters and cardiopulmonary function to provide a scientific basis for treatment;

[0119] Obtain an evaluation function, and input the first feature score and the second feature score into the evaluation function to obtain a correlation degree.

[0120] Obtain a preset evaluation function, and input the first feature score and the second feature score into the evaluation function to calculate the distance between the first feature type data and the second feature type data, and the measurement result is the correlation degree between the first feature type data and the second feature type data. The correlation degree is an important indicator for evaluating an individual's health level, which can reflect the coordination and consistency between the individual's exercise parameters and cardiopulmonary function, and can also perform dynamic monitoring and adjust the allocation and use efficiency of resources to reduce energy consumption.

[0121] Embodiment 2

[0122] See Figure 2 As shown, this embodiment provides a correlation analysis system for spinal-pelvic sagittal morphological parameters and cardiopulmonary exercise endurance parameters based on artificial intelligence, including:

[0123] A data acquisition module, which is used to obtain spinal-pelvic sagittal morphological parameters and cardiopulmonary exercise endurance parameters;

[0124] A preprocessing module, which is used to preprocess the spinal-pelvic sagittal morphological parameters, and the preprocessing includes classification and denoising operations;

[0125] A feature extraction module, which is used to construct an associated neural network, and input the preprocessed spinal-pelvic sagittal morphological parameters and cardiopulmonary exercise endurance parameters into the associated neural network for feature extraction, and output the first feature information of the spinal-pelvic sagittal morphological parameters and the second feature information of the cardiopulmonary exercise endurance parameters;

[0126] A correlation analysis module, which is used to obtain a preset correlation analysis model, and input the first feature information and the second feature information into the correlation analysis model to output the correlation degree between the spinal-pelvic sagittal morphological parameters and the cardiopulmonary exercise endurance parameters;

[0127] An evaluation module, which is used to obtain a preset evaluation threshold and compare the correlation degree with the evaluation threshold;

[0128] If the correlation degree is greater than the evaluation threshold, it indicates that there is a significant correlation between the spinal-pelvic sagittal morphological parameters and the cardiopulmonary exercise endurance parameters;

[0129] If the correlation degree is less than or equal to the evaluation threshold, it indicates that the correlation between the spinal-pelvic sagittal morphological parameters and the cardiopulmonary exercise endurance parameters is not significant.

[0130] Specifically, the data acquisition module includes:

[0131] The first acquisition unit, which is used to acquire the spinal-pelvic sagittal morphological parameters and the cardiopulmonary exercise endurance parameters.

[0132] Specifically, the preprocessing module includes:

[0133] The classification unit, which is used to classify the spinal-pelvic sagittal morphological parameters into training sample data and target analysis data according to the source of the spinal-pelvic sagittal morphological parameters. Among them, the target analysis data is used as the analysis basis for the correlation between the spinal-pelvic sagittal morphological parameters and the cardiopulmonary exercise endurance parameters in the follow-up;

[0134] The screening unit, which is used to determine the standard morphological parameters according to the training sample data, and then remove the abnormal data in the target analysis data according to the standard morphological parameters to obtain the normal data in the target analysis data;

[0135] Determine multiple sample morphological parameters according to the training sample data;

[0136] Determine the sample fluctuation range according to the multiple sample morphological parameters;

[0137] Obtain the sample evaluation function, and input the sample fluctuation range into the sample evaluation function for calculation to obtain multiple sample evaluation values;

[0138] Arrange the multiple sample evaluation values in descending order, and use the sample morphological parameter corresponding to the sample evaluation value with the highest ranking as the standard morphological parameter;

[0139] Obtain the screening range, and screen the target analysis data under the standard morphological parameters according to the screening range to obtain the normal data in the target analysis data.

[0140] The denoising unit, which is used to obtain the noise removal subset and perform denoising processing on the normal data in the noise removal subset to obtain the noise removal data;

[0141] Among them, performing denoising processing on the normal data in the noise removal subset to obtain the noise removal data includes:

[0142] Obtain the noise removal subset;

[0143] Determine a plurality of first adjacent parameters based on the normal data in the noise removal subset;

[0144] Determine the corresponding pre-bias amount according to the plurality of first adjacent parameters;

[0145] Obtain a denoising function, and input the pre-bias amount and each normal data into the denoising function to obtain noise-removed data.

[0146] An update unit, which is used to obtain a plurality of noise-removed data within a monitoring period, arrange them in chronological order to obtain a sequence to be evaluated, and then update the plurality of noise-removed data according to the sequence to be evaluated to obtain pre-parameters, and use the pre-parameters as the analysis basis for the correlation between the spinal-pelvic sagittal plane morphological parameters and the cardiopulmonary exercise endurance parameters;

[0147] Among them, the step of updating the plurality of noise-removed data according to the sequence to be evaluated to obtain pre-parameters includes:

[0148] Determine the sequence to be evaluated according to the noise-removed data;

[0149] Determine a plurality of second adjacent parameters according to the sequence to be evaluated;

[0150] Determine the corresponding post-bias amount according to the plurality of second adjacent parameters;

[0151] Obtain an update function, and input the post-bias amount and each noise-removed data into the update function to obtain a plurality of pre-parameters.

[0152] Specifically, the feature extraction module includes:

[0153] A feature extraction module, which is used to construct an association neural network, input the preprocessed spinal-pelvic sagittal plane morphological parameters and cardiopulmonary exercise endurance parameters into the association neural network for feature extraction, and output the first feature information of the spinal-pelvic sagittal plane morphological parameters and the second feature information of the cardiopulmonary exercise endurance parameters;

[0154] Among them, the feature extraction module includes:

[0155] An extraction unit, which is used to construct the association neural network and transfer the preprocessed spinal-pelvic sagittal plane morphological parameters and cardiopulmonary exercise endurance parameters as inputs to the input layer of the neural network;

[0156] A transfer unit, which is used to transfer the spinal-pelvic sagittal plane morphological parameters and cardiopulmonary exercise endurance parameters in the input layer to the hidden layer, and perform non-linear transformation and feature extraction in the hidden layer;

[0157] A fusion unit, which is used to fuse the features output by multiple hidden layers and transmit them to the output layer to obtain the first feature information of the spinal-pelvic sagittal plane morphological parameters and the second feature information of the cardiopulmonary exercise endurance parameters.

[0158] Specifically, the correlation analysis module includes:

[0159] A first input unit, which is used to obtain a preset correlation analysis model and input the first feature information and the second feature information into the correlation analysis model to obtain a first feature score corresponding to the first feature information and a second feature score corresponding to the second feature information;

[0160] A calculation unit, which is used to obtain an evaluation function and input the first feature score and the second feature score into the evaluation function to obtain a correlation degree.

[0161] The correlation analysis method for spinal-pelvic sagittal plane morphological parameters and cardiopulmonary exercise endurance parameters based on artificial intelligence provided by the present invention can more accurately determine the significant correlation between the spinal-pelvic sagittal plane morphological parameters and the cardiopulmonary exercise endurance parameters through data collection and processing, providing a scientific basis for clinical medicine and sports science. Through the introduction of a neural network, it brings a powerful feature extraction ability, provides a more accurate and robust data basis for correlation analysis, reduces the complexity of data processing and calculation cost, improves the analysis efficiency, can quickly process a large amount of medical data, provides real-time parameter evaluation for doctors and patients, and provides strong scientific support for training and rehabilitation in the field of sports medicine.

[0162] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. What is described in the above embodiments and the specification only illustrates the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of protection claimed by the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for analyzing the correlation between spine-pelvis sagittal morphological parameters and cardiopulmonary exercise endurance parameters based on artificial intelligence, characterized in that, Including: S1. Obtain spinal-pelvic sagittal morphological parameters and cardiopulmonary exercise endurance parameters; S2. Preprocess the spinal-pelvic sagittal morphological parameters; including: Classify the spinal-pelvic sagittal morphological parameters into training sample data and target analysis data according to their sources, where the target analysis data serves as the basis for analyzing the correlation between the spinal-pelvic sagittal morphological parameters and the cardiopulmonary exercise endurance parameters; Determine standard morphological parameters based on the training sample data, and then remove abnormal data in the target analysis data according to the standard morphological parameters to obtain normal data in the target analysis data; Obtain a noise removal subset, and perform denoising processing on the normal data in the noise removal subset to obtain noise removal data; Determine a sequence to be evaluated according to the noise removal data; Determine a plurality of second adjacent parameters based on the sequence to be evaluated; Determine the corresponding post-deviation amount based on the plurality of second adjacent parameters; Obtain an update function, and input the post-deviation amount and each noise removal data into the update function to obtain a plurality of pre-parameters; Use the pre-parameters as the basis for analyzing the correlation between the spinal-pelvic sagittal morphological parameters and the cardiopulmonary exercise endurance parameters; S3. Perform feature extraction based on a correlation neural network; specifically including: constructing a correlation neural network, and inputting the preprocessed spinal-pelvic sagittal morphological parameters and cardiopulmonary exercise endurance parameters into the correlation neural network for feature extraction, and outputting feature information, where the feature information includes first feature information for the spinal-pelvic sagittal morphological parameters and second feature information for the cardiopulmonary exercise endurance parameters; S4. Output a correlation degree based on a preset correlation analysis model and feature information: obtain a preset correlation analysis model, and input the first feature information and the second feature information into the correlation analysis model to output the correlation degree between the spinal-pelvic sagittal morphological parameters and the cardiopulmonary exercise endurance parameters; S5. Obtain a preset evaluation threshold, and compare the correlation degree with the evaluation threshold; the comparison criteria include: T1. If the correlation degree is greater than the evaluation threshold, it indicates that there is a significant correlation between the spinal-pelvic sagittal morphological parameters and the cardiopulmonary exercise endurance parameters; T2. If the correlation degree is less than or equal to the evaluation threshold, it indicates that the correlation between the spinal-pelvic sagittal morphological parameters and the cardiopulmonary exercise endurance parameters is not significant.

2. The correlation analysis method of spinal-pelvic sagittal morphological parameters and cardiopulmonary exercise endurance parameters based on artificial intelligence according to claim 1, characterized in that The step of determining standard morphological parameters based on the training sample data, and then removing abnormal data in the target analysis data according to the standard morphological parameters to obtain normal data in the target analysis data includes: Determine a plurality of sample morphological parameters based on the training sample data; Determine a sample fluctuation range based on the plurality of sample morphological parameters; Obtain a sample evaluation function, and input the sample fluctuation range into the sample evaluation function for calculation to obtain a plurality of sample evaluation values; Arrange the plurality of sample evaluation values in descending order, and use the sample morphological parameter corresponding to the sample evaluation value with the highest ranking as the standard morphological parameter; Obtain a screening interval, and screen the target analysis data under the standard morphological parameters according to the screening interval to obtain the normal data in the target analysis data.

3. The method for analyzing the correlation between spinal-pelvic sagittal morphological parameters and cardiopulmonary exercise endurance parameters based on artificial intelligence according to claim 2, wherein The step of determining the sample fluctuation interval according to the multiple sample morphological parameters includes: Obtain multiple sample morphological parameters; Determine the corresponding edge parameters according to the value length of each sample morphological parameter, where the sample morphological parameter is the same as the edge parameter; Obtain an edge measurement function, and input the edge measurement function and the edge parameter corresponding to the sample morphological parameter into the edge measurement function to obtain the sample fluctuation interval, where the number of sample fluctuation intervals is the same as the number of sample morphological parameters.

4. The method for analyzing the correlation between the spinal-pelvic sagittal morphological parameters and the cardiopulmonary exercise endurance parameters based on artificial intelligence according to claim 3, wherein The step of constructing an associated neural network, inputting the preprocessed spinal-pelvic sagittal morphological parameters and cardiopulmonary exercise endurance parameters into the associated neural network for feature extraction, and outputting the first feature information of the spinal-pelvic sagittal morphological parameters and the second feature information of the cardiopulmonary exercise endurance parameters includes: Construct an associated neural network, and use the preprocessed spinal-pelvic sagittal morphological parameters and cardiopulmonary exercise endurance parameters as inputs and transfer them to the input layer of the neural network; The neural network has multiple hidden layers, and transfer the spinal-pelvic sagittal morphological parameters and cardiopulmonary exercise endurance parameters in the input layer to the hidden layer, and perform non-linear transformation and feature extraction in the hidden layer; Fuse the features output by multiple hidden layers and transfer them to the output layer to obtain the first feature information of the spinal-pelvic sagittal morphological parameters and the second feature information of the cardiopulmonary exercise endurance parameters.

5. A correlation analysis system for spinal-pelvic sagittal morphological parameters and cardiopulmonary exercise endurance parameters based on artificial intelligence, which is used to implement the correlation analysis method for spinal-pelvic sagittal morphological parameters and cardiopulmonary exercise endurance parameters based on artificial intelligence according to any one of claims 1-4, characterized in that, It includes: A data acquisition module for obtaining spinal-pelvic sagittal morphological parameters and cardiopulmonary exercise endurance parameters; A preprocessing module for preprocessing the spinal-pelvic sagittal morphological parameters, and the preprocessing includes classification and denoising operations; A feature extraction module for constructing an associated neural network, inputting the preprocessed spinal-pelvic sagittal morphological parameters and cardiopulmonary exercise endurance parameters into the associated neural network for feature extraction, and outputting the first feature information of the spinal-pelvic sagittal morphological parameters and the second feature information of the cardiopulmonary exercise endurance parameters; A correlation analysis module for obtaining a preset correlation analysis model, inputting the first feature information and the second feature information into the correlation analysis model, and outputting the correlation degree between the spinal-pelvic sagittal morphological parameters and the cardiopulmonary exercise endurance parameters; An evaluation module for obtaining a preset evaluation threshold and comparing the correlation degree with the evaluation threshold; If the correlation degree is greater than the evaluation threshold, it indicates that there is a significant correlation between the spinal-pelvic sagittal morphological parameters and the cardiopulmonary exercise endurance parameters; If the correlation degree is less than or equal to the evaluation threshold, it indicates that the correlation between the spinal-pelvic sagittal morphological parameters and the cardiopulmonary exercise endurance parameters is not significant.

Images