A method and system for detecting vascular sclerosis parameters based on light-sensing sensors

Abstract

The present invention provides a method and system for detecting vascular sclerosis parameters using a light-sensitive sensor. The method comprises: acquiring and processing raw pulse wave signals through a preset wearable device sensor to determine pulse feature points; transmitting the pulse feature points to a preset big data analysis platform for analysis and calculation to determine vital sign data; transmitting the vital sign data to a preset artificial intelligence algorithm for analysis and processing to determine analysis results; using the analysis results to determine whether the subject under test has precursors to vascular sclerosis; regularly collecting and transmitting the analysis results to a preset simulation device to establish a pathological model of precursors to vascular sclerosis; and determining in real time whether the subject's vital sign data is abnormal based on the pathological model of precursors to vascular sclerosis. The system comprises a feature extraction module, a data analysis module, a pathological model establishment module, and a real-time monitoring module. The present invention can quickly query the wearer's basic information and issue early warnings to users with a high degree of match.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method and system for detecting vascular sclerosis parameters based on a light sensor. Background Art

[0002] Currently, due to the complex causes of diseases, lesions can only be detected with the help of certain medical equipment. Furthermore, only when abnormalities occur can the cause of the disease be determined through blood draws, electrocardiograms, and X-rays at a hospital. Vascular sclerosis is a series of pathological changes caused by diffuse atherosclerosis, stenosis, and small vessel occlusion, which reduces blood supply to the brain. However, early detection and timely notification of vascular sclerosis can help users take preventive measures to prevent long-term vascular blockages, which can lead to more serious life-threatening consequences. Current vascular monitoring devices are relatively simple in functionality and lack intelligent capabilities, failing to intelligently monitor key vascular data.

[0003] Prior art 1, application number CN202110009183.3, discloses a device and method for measuring vascular sclerosis parameters, comprising: a geometric measurement module for acquiring a cross-sectional image and a longitudinal cross-sectional image of a blood vessel through an imaging device; a pressure measurement module for acquiring a pressure signal of the blood vessel through a sensor; a data processing module for analyzing the cross-sectional image and the longitudinal cross-sectional image to acquire a geometric change waveform, analyzing the pressure signal to acquire a pressure change waveform, and processing the geometric change waveform and the pressure change waveform to acquire a vascular sclerosis parameter; and a display output module for displaying the vascular sclerosis parameter. Although stable vascular sclerosis parameters can be obtained and have good clinical application value, the lack of relevant vascular sclerosis models to process abnormal parameters leads to inaccurate vascular sclerosis parameters, and no adjustment strategy for abnormal parameters is implemented.

[0004] Prior art 2, application number CN202210869321.X, discloses a method, device, and electronic device for assessing vascular sclerosis, including: acquiring Korotkoff sound signals from a blood vessel to be tested; performing scattering feature recognition on the Korotkoff sound signals to obtain a scattering feature set; dividing the scattering feature set into a training set and a test set according to a preset ratio; training a vascular sclerosis classification model based on the training set; and inputting the test set into the classification model to obtain a vascular sclerosis assessment result. Although the overall data processing efficiency is high and the implementation method is simple, and training the classification model can improve classification efficiency and accuracy, using only the vascular sclerosis classification model to analyze the signals in the test set results in inaccurate vascular sclerosis analysis results and has poor intelligent capabilities.

[0005] Prior art three, application number CN201510819873.X, discloses an arterial vascular sclerosis measurement device and charging base, comprising a cuff and a main unit connected by a gas conduit; the cuff contains one or two air bags; the main unit contains a controller and one or two inflation and deflation components, each of which includes a pressure sensor, an air pump, and an exhaust valve; the controller extracts one or more features from the pulse wave waveform obtained by the pressure sensor, which are used individually or in combination to determine the degree of arterial vascular sclerosis. While it can obtain a highly accurate pulse wave waveform, its structure is relatively simple and its functions are relatively limited, making it difficult to obtain accurate vascular sclerosis parameters.

[0006] Currently, the existing technologies 1, 2 and 3 have relatively simple functions and low intelligence levels. Therefore, the present invention provides a method and system for detecting vascular sclerosis parameters based on a light sensor. By setting up data monitoring, the patient's blood pressure, blood sugar, respiration and body temperature can be measured and monitored at any time, making data collection more complete, which is helpful to timely detect the rehabilitation and conditioning of vascular sclerosis. Summary of the Invention

[0007] In order to solve the above technical problems, the present invention provides a method for detecting vascular sclerosis parameters based on a light sensor, comprising the following steps:

[0008] Through the preset wearable device sensor, the original pulse wave signal is acquired and processed to determine the pulse feature points;

[0009] The pulse characteristic points are transmitted to a preset big data analysis platform for analysis and calculation to determine the vital sign data, which is then transmitted to a preset artificial intelligence algorithm for analysis and processing to determine the analysis results. The analysis results are used to determine whether the subject to be tested has signs of vascular sclerosis.

[0010] Based on a deep neural network learning algorithm, a pathological model of the precursors of vascular sclerosis is established by regularly collecting and analyzing results and transmitting them to a preset simulation device.

[0011] According to the pathological model of precursors to vascular sclerosis, it is possible to determine in real time whether the vital sign data of the subject to be tested is abnormal, and to issue an abnormal reminder when the vital sign data is abnormal.

[0012] Optionally, the process of determining the pulse feature points includes the following steps:

[0013] Through the preset wearable device sensor, the pulse sign information of the subject to be tested is detected regularly;

[0014] Determining the original pulse wave signal of the subject to be detected according to the pulse sign information;

[0015] Denoising, baseline removal and wavelet decomposition are performed on the original pulse wave signal to determine the preprocessed pulse wave signal;

[0016] According to the preprocessed pulse wave signal, pulse wave characteristic signal points are collected and reconstructed to determine the pulse wave characteristic points.

[0017] Optionally, the process of performing noise reduction, baseline removal, and wavelet decomposition on the original pulse wave signal includes the following steps:

[0018] Input the noisy original pulse wave signal, set the number of iterations, input the intermediate variables, perform total variation denoising on the noisy original pulse wave signal until it converges to the optimal value of the cost function. The iteration ends and the denoised original pulse wave signal ecg is obtained. 1 , length is L;

[0019] Select the window width W, W is an odd number, for ecg 1 The two ends of the extension are extended to obtain the original pulse wave signal ecg after extension 2 ;

[0020] Original pulse wave signal ecg 3 Perform wavelet decomposition.

[0021] Optionally, the process of transmitting the vital sign data to a preset artificial intelligence algorithm for analysis and processing includes the following steps:

[0022] Obtain pulse wave feature points and transmit them to a big data analysis platform;

[0023] Based on the big data analysis platform, the incoming pulse wave signal characteristics are analyzed and calculated to obtain vital sign data; wherein the vital sign data includes at least heart rate data, blood pressure data and blood oxygen data;

[0024] Through the preset artificial intelligence algorithm, all frequent data are first found from the vital sign data, and the support of the frequent data is greater than or equal to the minimum support threshold; association rules are generated from the frequent data, the confidence is calculated, and the association rules with confidence greater than or equal to the minimum confidence threshold are retained to complete the vital sign data mining and screening of the subjects to be tested, introduce corresponding treatment plans, and determine the analysis results.

[0025] Optionally, the process of processing the analysis results using the vascular sclerosis precursor pathological model includes the following steps:

[0026] Regularly collect and obtain analysis results, including: heart rate data, blood pressure data, and blood oxygen data of the subject to be tested;

[0027] Based on the analysis results, calculate the predicted analysis results in the simulation device and determine the difference between the analysis results and the predicted analysis results:

[0028] Evaluate the difference and determine the evaluation result:

[0029] When the mean absolute error, root mean square error, and mean relative error all meet the preset thresholds, the evaluation result is determined to be 1, and a pathological model of the precursors of vascular sclerosis is established based on a deep neural network learning algorithm;

[0030] When the mean absolute error, root mean square error, and mean relative error do not meet the preset thresholds, the evaluation result value is determined to be 0, indicating that the correlation between the historical analysis results and the real-time analysis results is too small, and an abnormal prompt is issued.

[0031] Optionally, the process of determining the vital sign data of the subject to be detected in real time includes the following steps:

[0032] Obtaining a timing acquisition frequency, and collecting state vital sign data of the subject to be detected according to the timing acquisition frequency;

[0033] Obtain historical vital sign data and calculate the variance fluctuation value of state vital sign data and historical state data;

[0034] Based on the pathological model of precursors to vascular sclerosis, the variance fluctuation value is used to determine whether the health status data of the subject to be tested is abnormal, and a judgment result is generated;

[0035] When the judgment result is abnormal, the pre-set reminder device is triggered, and a prompt is sent to the terminal device pre-bound to the wearable device sensor, and the data is continuously tracked and stored in the database; when the judgment result is normal, the status vital sign data is stored in the preset storage database.

[0036] The present invention provides a system for detecting vascular sclerosis parameters based on a light sensor, comprising:

[0037] The feature extraction module is responsible for acquiring and processing the original pulse wave signal through the preset smart device sensor and determining the pulse feature points;

[0038] The data analysis module is responsible for transmitting the pulse feature points to the preset big data analysis platform for analysis and calculation, determining the vital sign data, and transmitting the vital sign data to the preset artificial intelligence algorithm for analysis and processing to determine the analysis results, which are used to determine whether the subject to be tested has signs of vascular sclerosis;

[0039] Establish a pathological model module, which is responsible for establishing a pathological model of the precursors of vascular sclerosis by regularly collecting and analyzing results and transmitting them to a preset simulation device based on a deep neural network learning algorithm;

[0040] The real-time monitoring module is responsible for judging whether the vital sign data of the subject to be tested is abnormal in real time based on the precursor pathological model of vascular sclerosis. When the vital sign data is abnormal, it will issue an abnormal reminder and make corresponding maintenance strategies in time.

[0041] Optional feature extraction module, including:

[0042] The original pulse wave signal acquisition submodule is responsible for regularly detecting the pulse sign information of the subject to be detected through the preset wearable device sensor, and determining the original pulse wave signal of the subject to be detected based on the pulse sign information;

[0043] The original pulse wave signal processing submodule is responsible for performing noise reduction, baseline removal and wavelet decomposition on the original pulse wave signal to determine the preprocessed pulse wave signal;

[0044] The pulse wave feature point reconstruction submodule is responsible for collecting and reconstructing the pulse wave feature signal points based on the preprocessed pulse wave signal, and determining the pulse wave feature points.

[0045] Optional data analysis module, including:

[0046] The transmission submodule is responsible for obtaining pulse wave feature points and transmitting them to the big data analysis platform;

[0047] The parsing submodule is responsible for analyzing and calculating the vital sign data by analyzing the incoming pulse wave signal characteristics based on the big data analysis platform. The vital sign data includes at least heart rate data, blood pressure data, and blood oxygen data.

[0048] The analysis unit submodule is responsible for mining and screening the vital sign data of the test subject through a preset artificial intelligence algorithm, launching corresponding treatment plans, and determining the analysis results.

[0049] Optionally, establish a pathology model module, including:

[0050] The regular collection submodule is responsible for regular collection and obtaining analysis results;

[0051] The difference calculation submodule is responsible for calculating the predicted analysis results in the simulation device based on the analysis results and determining the difference between the analysis results and the predicted analysis results;

[0052] The evaluation submodule is responsible for evaluating the difference and determining the evaluation results;

[0053] A pathology model submodule is established, which is responsible for determining the evaluation result value to be 1 when the mean absolute error, root mean square error, and mean relative error all meet the preset thresholds. Based on the deep neural network learning algorithm, a pathology model for the precursor of vascular sclerosis is established;

[0054] The abnormality prompt submodule is responsible for determining that the value of the evaluation result is 0 when the mean absolute error, root mean square error, and mean relative error do not meet the preset thresholds, indicating that the correlation between the historical analysis results and the real-time analysis results is too small, and an abnormality prompt is issued.

[0055] The present invention acquires and processes the original pulse wave signal through a preset wearable device sensor to determine the pulse characteristic points; secondly, the pulse characteristic points are transmitted to a preset big data analysis platform for analytical calculation to determine the vital sign data, and the vital sign data are transmitted to a preset artificial intelligence algorithm for analysis and processing to determine the analysis results; the analysis results are used to determine whether the subject to be detected has precursors of vascular sclerosis; then, based on a deep neural network learning algorithm, the analysis results are regularly collected and transmitted to a preset simulation device to establish a precursor pathological model of vascular sclerosis; finally, based on the precursor pathological model of vascular sclerosis, it is determined in real time whether the vital sign data of the subject to be detected is abnormal, and an abnormal reminder is issued when the vital sign data is abnormal; the above scheme extracts pulse wave signal features based on the pulse wave data of the human body, and the smart wearable device analyzes the wearer's physical health indicators Real-time detection, and comparison of the heart rate, blood pressure and blood oxygen data obtained from the detection with the characteristics of physiological indicators related to vascular sclerosis, and early warning of eligible situations, so as to inform the family or hospital immediately when the wearer's physical indicators are abnormal, thus avoiding the user's body from being in danger of life due to uncontrollable factors; through the intelligent early warning of vascular sclerosis parameters, the wearer's basic information can be quickly queried, and users with a high degree of matching can be warned, and professional preliminary physiological data analysis can be given, which is convenient for doctors and patients to conduct early warning, mid-term control and late treatment of certain symptoms, complications and risk factors. By setting up resource sharing, it is convenient for patients to measure and monitor the patient's blood pressure, blood sugar, respiration and body temperature at any time through data monitoring, making data collection more complete and helping to timely detect rehabilitation and conditioning of vascular sclerosis.

[0056] Other features and advantages of the present invention will be described in the following description, and in part will become apparent from the description, or will be understood by practicing the present invention. The purposes and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings.

[0057] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0059] Figure 1This is a flow chart of a method for detecting vascular sclerosis parameters based on a light sensor in Example 1 of the present invention;

[0060] Figure 2 FIG. 1 is a diagram showing a process for determining pulse characteristic points in Example 2 of the present invention;

[0061] Figure 3 1 is a diagram showing the process of performing noise reduction, baseline removal, and wavelet decomposition on the original pulse wave signal in Example 3 of the present invention;

[0062] Figure 4 This is a diagram of the process of transmitting vital sign data to a preset artificial intelligence algorithm for analysis and processing in Example 4 of the present invention;

[0063] Figure 5 FIG. 1 is a diagram illustrating the processing of analysis results by the pathological model for precursors of vascular sclerosis in Example 5 of the present invention;

[0064] Figure 6 This is a diagram of the process of determining the vital sign data of the subject to be detected in real time in Example 6 of the present invention;

[0065] Figure 7 This is a block diagram of a system for detecting vascular sclerosis parameters based on a light sensor in Example 7 of the present invention;

[0066] Figure 8 This is a block diagram of a feature extraction module in Example 8 of the present invention;

[0067] Figure 9 This is a block diagram of the data analysis module in Example 9 of the present invention;

[0068] Figure 10 This is a block diagram of a pathology model module in Example 10 of the present invention;

[0069] Figure 11 This is a block diagram of the real-time monitoring module in Example 11 of the present invention. DETAILED DESCRIPTION

[0070] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0071] The terms used in the embodiments of the present application are for the purpose of describing specific embodiments only and are not intended to limit the embodiments of the present application. The singular forms "a," "the," and "the" used in the embodiments of the present application and the appended claims are also intended to include plural forms unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0072] When the following description refers to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims. In the description of the present application, it should be understood that the terms "first", "second", "third", etc. are only used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence, nor can they be understood as indicating or implying relative importance. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to the specific circumstances.

[0073] Example 1: Figure 1 As shown, an embodiment of the present invention provides a method for detecting vascular sclerosis parameters based on a light sensor, comprising the following steps:

[0074] S100: Obtaining and processing the original pulse wave signal through a preset wearable device sensor to determine the pulse feature points;

[0075] S200: transmitting the pulse characteristic points to a preset big data analysis platform for analysis and calculation to determine vital sign data, and transmitting the vital sign data to a preset artificial intelligence algorithm for analysis and processing to determine analysis results; the analysis results are used to determine whether the subject to be tested has precursors to vascular sclerosis;

[0076] S300: Based on a deep neural network learning algorithm, it establishes a pathological model of the precursors of vascular sclerosis by regularly collecting and analyzing results and transmitting them to a pre-set simulation device.

[0077] S400: judging in real time whether the physical sign data of the subject to be tested is abnormal based on the pathological model of precursors to vascular sclerosis, and issuing an abnormal reminder if the physical sign data is abnormal;

[0078] The working principle and beneficial effects of the above technical solution are as follows: this embodiment first obtains and processes the original pulse wave signal through the preset wearable device sensor to determine the pulse feature points; secondly, the pulse feature points are transmitted to the preset big data analysis platform for analysis and calculation to determine the vital sign data, and the vital sign data are transmitted to the preset artificial intelligence algorithm for analysis and processing to determine the analysis results; the analysis results are used to determine whether the object to be detected has precursors of vascular sclerosis; then, based on the deep neural network learning algorithm, the analysis results are regularly collected and transmitted to the preset simulation equipment to establish a precursor pathological model of vascular sclerosis; finally, according to the precursor pathological model of vascular sclerosis, it is judged in real time whether the vital sign data of the object to be detected is abnormal, and an abnormal reminder is issued when the vital sign data is abnormal; the above solution extracts pulse wave signal features based on the pulse wave data of the human body, and the smart wearable device performs pulse wave signal extraction on the wearer. The wearer's physical health indicators are detected in real time, and the heart rate, blood pressure and blood oxygen data obtained from the detection are compared with the characteristics of the physiological indicators related to vascular sclerosis, and early warnings are issued when the conditions are met, so that family members or hospitals can be informed immediately when the wearer's physical indicators are abnormal, thereby avoiding the user's body from being in danger of life due to uncontrollable factors; through the intelligent early warning of vascular sclerosis parameters, the wearer's basic information can be quickly queried, and users with a high degree of matching can be warned, and professional preliminary physiological data analysis can be given, which is convenient for doctors and patients to carry out early warning, mid-term control and late treatment of certain symptoms, complications and risk factors. By setting up resource sharing, it is convenient for patients to measure and monitor the patient's blood pressure, blood sugar, respiration and body temperature at any time through data monitoring, making data collection more complete and helping to timely discover rehabilitation and conditioning of vascular sclerosis.

[0079] Example 2: Figure 2 As shown, based on Example 1, the pulse characteristic point determination process provided by the embodiment of the present invention includes the following steps:

[0080] S101: Regularly detecting pulse and vital signs of a subject through a preset wearable device sensor;

[0081] S102: Determine the original pulse wave signal of the subject to be detected based on the pulse sign information;

[0082] S103: performing noise reduction, baseline removal, and wavelet decomposition on the original pulse wave signal to determine a preprocessed pulse wave signal;

[0083] S104: collecting and reconstructing pulse wave characteristic signal points based on the preprocessed pulse wave signal to determine pulse wave characteristic points;

[0084] The working principle and beneficial effects of the above technical solution are as follows: first, the present embodiment detects the pulse vital signs information of the subject to be detected by a preset wearable device sensor at regular intervals; secondly, the original pulse wave signal of the subject to be detected is determined based on the pulse vital signs information; then, the original pulse wave signal is subjected to noise reduction, baseline removal and wavelet decomposition to determine a pre-processed pulse wave signal; finally, based on the pre-processed pulse wave signal, the pulse wave characteristic signal points are collected and reconstructed to determine the pulse wave characteristic points; the above solution uses a wearable device sensor to realize the collection of pulse vital signs information, which is firstly very convenient to use and can be realized through a wearable device and is very easy to carry; secondly, the wearable device realizes the real-time collection of pulse characteristic information, which is helpful The accuracy of the results of vascular sclerosis parameter analysis is guaranteed; by performing noise reduction, baseline removal and wavelet decomposition on the original pulse wave signal, important information of the pulse wave feature points is extracted from the original pulse wave signal through noise reduction, and other electromagnetic interference is removed, effectively ensuring the accuracy of the pulse wave feature points. By removing the baseline, the baseline removal processing is achieved, and the similarity of the pulse wave feature points with the original pulse wave signal is guaranteed. The original pulse wave signal will not be distorted due to the baseline removal processing. The original pulse wave signal is decomposed into different components through wavelet decomposition, and then the target component is retained and the non-target component is suppressed. The original pulse wave signal is then reconstructed to obtain the filtered original pulse wave signal, and finally the preprocessed pulse wave signal is obtained.

[0085] Example 3: Figure 3 As shown, based on Example 2, the process of performing noise reduction, baseline removal, and wavelet decomposition on the original pulse wave signal provided by the embodiment of the present invention includes the following steps:

[0086] S1031: Input the original noisy pulse wave signal, set the number of iterations, input the intermediate variables, and perform total variation denoising on the original noisy pulse wave signal until it converges to the optimal value of the cost function. The iteration ends and the denoised original pulse wave signal ecg is obtained. 1 , length is L;

[0087] S1032: Select the window width W, W is an odd number, for ecg 1 The two ends of the extension are extended to obtain the original pulse wave signal ecg after extension 2 , the processing method is as follows:

[0088]

[0089] Where i represents the identification value of the original pulse wave signal after extension, and the original pulse wave signal ecg after extension 2 The length is L+W-2;

[0090] ECG 2Add window and perform median filtering on the original pulse wave signal in the window, that is, sort the original pulse wave signal in the window, replace the value of the window center point with the median, move the window, and traverse the ecg 2 , fitting the drifted baseline BL, BL and ecg 2 The relational expression is:

[0091] BL=median[ecg 2 (i):ecg 2 (i+W)],0≤i≤L-1

[0092] Among them, the median() function represents the median operation, from ecg 1 Subtract BL from the original pulse wave signal ecg after eliminating baseline drift 3 ;

[0093] S1033: original pulse wave signal ecg 3 Perform wavelet decomposition, and the expression of the wavelet decomposition coefficient is:

[0094]

[0095] in, represents the wavelet estimation coefficient of the denoised signal, ω represents the wavelet coefficient of the original signal, a represents the adjustment factor of the wavelet decomposition, and λ represents the wavelet coefficient threshold;

[0096] The working principle and beneficial effects of the above technical solution are as follows: In this embodiment, the original pulse wave signal containing noise is first input, the number of iterations is set, the intermediate variables are input, and the total variation denoising is performed on the original pulse wave signal containing noise until it converges to the optimal value of the cost function. The iteration ends and the original pulse wave signal ecg after denoising is obtained. 1 , length is L; secondly, select the window width W, W is an odd number, for ecg 1 The two ends of the extension are extended to obtain the original pulse wave signal ecg after extension 2 , from ecg 1 Subtract BL from the original pulse wave signal ecg after eliminating baseline drift 3 ; Finally, the original pulse wave signal ecg 3 Perform wavelet decomposition; the above scheme uses noise reduction, baseline removal and wavelet decomposition to gradually process the original pulse wave signal, which on the one hand effectively ensures the accuracy of the pulse wave feature points, and on the other hand ensures the similarity between the pulse wave feature points and the original pulse wave signal, suppresses non-target components, and ensures the accuracy of the preprocessed pulse wave signal.

[0097] Example 4: Figure 4As shown, the process of transmitting vital sign data to a preset artificial intelligence algorithm for analysis and processing provided by an embodiment of the present invention includes the following steps:

[0098] S201: Acquire pulse wave feature points and transmit the pulse wave feature points to a big data analysis platform;

[0099] S202: Analyzing the characteristics of the incoming pulse wave signal based on the big data analysis platform, and analyzing and calculating to obtain vital sign data; wherein the vital sign data includes at least heart rate data, blood pressure data, and blood oxygen data;

[0100] S203: Using a preset artificial intelligence algorithm, first find all frequent data from the vital sign data, where the support of the frequent data is greater than or equal to a minimum support threshold; generate association rules from the frequent data, calculate confidence, and retain association rules with confidence greater than or equal to the minimum confidence threshold, thereby completing the mining and screening of the vital sign data of the subject to be tested, proposing a corresponding treatment plan, and determining the analysis results;

[0101] The working principle and beneficial effects of the above technical solution are as follows: This embodiment first obtains pulse wave feature points and transmits them to a big data analysis platform; secondly, based on the big data analysis platform, the incoming pulse wave signal features are analyzed and calculated to obtain vital sign data; wherein the vital sign data includes at least heart rate data, blood pressure data, and blood oxygen data; finally, a preset artificial intelligence algorithm is used to first identify all frequent data from the vital sign data, where the support of the frequent data is greater than or equal to a minimum support threshold; association rules are generated from the frequent data, confidence levels are calculated, and association rules with confidence levels greater than or equal to the minimum confidence threshold are retained, thereby completing the mining and screening of the vital sign data of the subject to be tested, proposing a corresponding treatment plan, and determining the analysis results; the above solution analyzes the pulse wave feature points through the big data analysis platform, realizes the identification of vital sign data, and lays a data foundation for the mining and screening of vital sign data; the artificial intelligence algorithm is used to mine and screen the vital sign data, realizing intelligent processing of the vital sign data, and by setting association rules, completes the mining and screening of preset vital sign data, ensuring the integrity of the vital sign data.

[0102] Example 5: Figure 5 As shown, based on Example 1, the process of processing the analysis results of the vascular sclerosis precursor pathological model provided in the embodiment of the present invention includes the following steps:

[0103] S301: Regularly collect and obtain analysis results, the analysis results including: heart rate data, blood pressure data and blood oxygen data of the subject to be tested;

[0104] S302: Calculate the predicted analysis result in the simulation device based on the analysis result, and determine the difference between the analysis result and the predicted analysis result:

[0105] S303: Evaluate the difference and determine the evaluation result:

[0106] S304: When the mean absolute error, the root mean square error, and the mean relative error all meet preset thresholds, determining the value of the evaluation result to be 1, and establishing a pathological model for the precursor of vascular sclerosis based on a deep neural network learning algorithm;

[0107] S305: When the mean absolute error, the root mean square error, and the mean relative error do not meet the preset thresholds, the evaluation result is determined to be 0, indicating that the correlation between the historical analysis results and the real-time analysis results is too low, and an abnormality prompt is issued;

[0108] The working principle and beneficial effects of the above technical solution are as follows: This embodiment first regularly collects and obtains analysis results, and the analysis results include: heart rate data, blood pressure data and blood oxygen data of the subject to be detected; according to the analysis results, the predicted analysis results in the simulation device are calculated, and the difference between the analysis results and the predicted analysis results is determined; the difference is evaluated to determine the evaluation result: when the mean absolute error, root mean square error and mean relative error all meet the preset threshold, the value of the evaluation result is determined to be 1, and based on the deep neural network learning algorithm, a precursor pathological model of vascular sclerosis is established; when the mean absolute error, root mean square error and mean relative error do not meet the preset threshold, the evaluation result is determined to be 1. The value of the estimated result is 0, which means that the correlation between the historical analysis results and the real-time analysis results is too small, and an abnormal prompt is given; the above scheme analyzes and processes the analysis results through the model, and obtains parameters with a high degree of correlation with vascular sclerosis, ensuring the accuracy of the vascular sclerosis degree results, eliminating the tediousness of separately collecting heart rate, blood pressure and blood oxygen, shortening the inspection time, and truly realizing the intelligent analysis of pulse wave signals; at the same time, the vascular sclerosis precursor pathology model compares the root mean square error and the mean relative error with the preset threshold value, and the evaluation result is determined by these two factors, which makes up for the inaccuracy of single factor measurement and further improves the accuracy of vascular sclerosis.

[0109] Example 6: Figure 6 As shown, based on Example 1, the process of real-time determination of vital sign data of a subject to be detected provided by the embodiment of the present invention includes the following steps:

[0110] S401: Acquire a timing acquisition frequency, and acquire state vital sign data of the subject to be detected according to the timing acquisition frequency;

[0111] S402: Acquire historical vital sign data and calculate the variance fluctuation value of the state vital sign data and the historical state data;

[0112] S403: Based on the pathological model of precursors to vascular sclerosis, judging whether the health status data of the subject to be tested is abnormal by using the variance fluctuation value, and generating a judgment result;

[0113] S404: When the judgment result is abnormal, the pre-set reminder device is triggered, and a reminder is sent to the terminal device pre-bound to the wearable device sensor, and the data is continuously tracked and stored; when the judgment result is normal, the status vital sign data is stored in a preset storage database;

[0114] The working principle and beneficial effects of the above technical solution are as follows: This embodiment first obtains a timed acquisition frequency and, based on the timed acquisition frequency, collects the status vital sign data of the subject to be detected; secondly, obtains historical vital sign data and calculates the variance fluctuation value of the status vital sign data and the historical status data; then, based on the pathological model of the precursor of vascular sclerosis, the variance fluctuation value is used to determine whether the physical health status data of the subject to be detected is abnormal, and a judgment result is generated; finally, when the judgment result is abnormal, a pre-set reminder device is triggered, and a prompt is sent to the terminal device pre-bound to the wearable device sensor, and the data is continuously tracked and stored; when the judgment result is normal, the status vital sign data is stored in a preset storage database; the above solution uses the variance fluctuation value to determine whether the physical health status data of the subject to be detected is abnormal. By comparing the data, the accuracy of the judgment can be greatly improved, and it is more precise, eliminating the tedious traditional technology of doctors' judgment based on experience, saving medical resources, and improving the intelligence of data processing. The comparison of data also facilitates the intuitive understanding of the subject to be detected and improves the user experience of the subject to be detected.

[0115] Example 7: Figure 7 As shown, based on Examples 1 to 6, the system for detecting vascular sclerosis parameters based on a light sensor according to an embodiment of the present invention includes:

[0116] The feature extraction module is responsible for acquiring and processing the original pulse wave signal through the preset smart device sensor and determining the pulse feature points;

[0117] The data analysis module is responsible for transmitting the pulse feature points to the preset big data analysis platform for analysis and calculation, determining the vital sign data, and transmitting the vital sign data to the preset artificial intelligence algorithm for analysis and processing to determine the analysis results, which are used to determine whether the subject to be tested has signs of vascular sclerosis;

[0118] Establish a pathological model module, which is responsible for establishing a pathological model of the precursors of vascular sclerosis by regularly collecting and analyzing results and transmitting them to a preset simulation device based on a deep neural network learning algorithm;

[0119] The real-time monitoring module is responsible for determining whether the subject's vital signs are abnormal in real time based on the pathological model of vascular sclerosis precursors. If the vital signs are abnormal, it will issue an abnormal reminder and formulate corresponding treatment strategies in a timely manner.

[0120] The working principle and beneficial effects of the above technical solution are as follows: the feature extraction module of this embodiment obtains and processes the original pulse wave signal through the preset intelligent device sensor to determine the pulse feature points; the data analysis module transmits the pulse feature points to the preset big data analysis platform and performs analytical calculations to determine the vital sign data, and transmits the vital sign data to the preset artificial intelligence algorithm for analysis and processing to determine the analysis results, which are used to determine whether the object to be detected has precursors of vascular sclerosis; the pathological model establishment module establishes a vascular sclerosis model based on the deep neural network learning algorithm by regularly collecting analysis results and transmitting them to the preset simulation equipment; the real-time monitoring module determines whether the vital sign data of the object to be detected is abnormal in real time based on the pathological model of precursors of vascular sclerosis, and when the vital sign data is abnormal, it issues an abnormal reminder and promptly formulates corresponding conditioning strategies; the above solution performs pulse wave signal feature extraction based on the pulse wave data of the human body Smart wearable devices can detect the wearer's physical health indicators in real time, and compare the heart rate, blood pressure and blood oxygen data obtained from the detection with the characteristics of physiological indicators related to vascular sclerosis, and issue early warnings when the conditions are met, so as to inform the family or hospital immediately when the wearer's physical indicators are abnormal, thus avoiding the user's life-threatening due to uncontrollable factors; through the intelligent early warning of vascular sclerosis parameters, the wearer's basic information can be quickly queried, and users with a high degree of matching can be warned, and professional preliminary physiological data analysis can be given, so that doctors and patients can conduct early warning, mid-term control and late treatment of certain symptoms, complications and risk factors. By setting up resource sharing, patients can conveniently measure and monitor the patient's blood pressure, blood sugar, respiration and body temperature at any time through data monitoring, making data collection more complete and helping to timely detect rehabilitation and conditioning of vascular sclerosis.

[0121] Example 8: Figure 8 As shown, based on Example 7, the feature extraction module provided by the embodiment of the present invention includes:

[0122] The original pulse wave signal acquisition submodule is responsible for regularly detecting the pulse sign information of the subject to be detected through the preset wearable device sensor, and determining the original pulse wave signal of the subject to be detected based on the pulse sign information;

[0123] The original pulse wave signal processing submodule is responsible for performing noise reduction, baseline removal and wavelet decomposition on the original pulse wave signal to determine the preprocessed pulse wave signal;

[0124] The pulse wave feature point reconstruction submodule is responsible for collecting and reconstructing the pulse wave feature signal points based on the preprocessed pulse wave signal and determining the pulse wave feature points;

[0125] The working principle and beneficial effects of the above technical solution are as follows: the original pulse wave signal acquisition submodule of this embodiment detects the pulse sign information of the object to be detected regularly through the preset wearable device sensor, and determines the original pulse wave signal of the object to be detected based on the pulse sign information; the original pulse wave signal processing submodule performs noise reduction, baseline removal and wavelet decomposition on the original pulse wave signal to determine the preprocessed pulse wave signal; the pulse wave feature point reconstruction submodule collects and reconstructs the pulse wave feature signal points based on the preprocessed pulse wave signal to determine the pulse wave feature points; the above solution uses the wearable device sensor to realize the collection of pulse sign information, which is firstly very convenient to use, can be realized through the wearable device, and is very easy to carry, and secondly, the pulse feature points are realized by the wearable device The real-time collection of characteristic information helps to ensure the accuracy of the results of vascular sclerosis parameter analysis; by performing noise reduction, baseline removal and wavelet decomposition on the original pulse wave signal, important information of the pulse wave feature points is extracted from the original pulse wave signal through noise reduction, and other electromagnetic interference is removed, effectively ensuring the accuracy of the pulse wave feature points. By removing the baseline, the baseline removal processing is achieved, and the similarity of the pulse wave feature points with the original pulse wave signal is guaranteed, and the original pulse wave signal will not be distorted due to the baseline removal processing. The original pulse wave signal is decomposed into different components through wavelet decomposition, and then the target component is retained and the non-target component is suppressed. The original pulse wave signal is then reconstructed to obtain the filtered original pulse wave signal, and finally the preprocessed pulse wave signal is obtained.

[0126] Example 9: Figure 9 As shown, based on Example 7, the data analysis module provided by this embodiment of the present invention includes:

[0127] The transmission submodule is responsible for obtaining pulse wave feature points and transmitting them to the big data analysis platform;

[0128] The parsing submodule is responsible for analyzing and calculating the vital sign data by analyzing the incoming pulse wave signal characteristics based on the big data analysis platform. The vital sign data includes at least heart rate data, blood pressure data, and blood oxygen data.

[0129] The analysis unit submodule is responsible for mining and screening the vital sign data of the test subject through a preset artificial intelligence algorithm, launching a corresponding treatment plan, and determining the analysis results;

[0130] The working principle and beneficial effects of the above technical solution are as follows: the transmission submodule of this embodiment obtains pulse wave characteristic points and transmits the pulse wave characteristic points to the big data analysis platform; the analysis submodule is based on the big data analysis platform, and analyzes and calculates the vital sign data by analyzing the incoming pulse wave signal characteristics; wherein the vital sign data at least includes heart rate data, blood pressure data and blood oxygen data; the analysis unit submodule mines and screens the vital sign data of the subject to be tested through a preset artificial intelligence algorithm, proposes a corresponding treatment plan, and determines the analysis results; the above solution analyzes the pulse wave characteristic points through the big data analysis platform, realizes the identification of vital sign data, and lays a data foundation for the mining and screening of vital sign data; through the artificial intelligence algorithm, the vital sign data is mined and screened, and the intelligent processing of the vital sign data is realized. Through the setting of association rules, the mining and screening of the preset required vital sign data is completed, and the integrity of the vital sign data is ensured.

[0131] Example 10: Figure 10 As shown, based on Example 7, the pathology model establishment module provided by the embodiment of the present invention includes:

[0132] The regular collection submodule is responsible for regular collection and obtaining analysis results;

[0133] The difference calculation submodule is responsible for calculating the predicted analysis results in the simulation device based on the analysis results and determining the difference between the analysis results and the predicted analysis results;

[0134] The evaluation submodule is responsible for evaluating the difference and determining the evaluation results;

[0135] A pathology model submodule is established, which is responsible for determining the evaluation result value to be 1 when the mean absolute error, root mean square error, and mean relative error all meet the preset thresholds. Based on the deep neural network learning algorithm, a pathology model for the precursor of vascular sclerosis is established;

[0136] The abnormality prompt submodule is responsible for determining the evaluation result value to be 0 when the mean absolute error, root mean square error, and mean relative error do not meet the preset thresholds, indicating that the correlation between the historical analysis results and the real-time analysis results is too small, and an abnormality prompt is issued;

[0137] The working principle and beneficial effects of the above technical solution are as follows: the periodic acquisition submodule of this embodiment regularly acquires and obtains analysis results; the difference calculation submodule calculates the predicted analysis results in the simulation device based on the analysis results, and determines the difference between the analysis results and the predicted analysis results; the evaluation submodule evaluates the difference and determines the evaluation result; the pathological model establishment submodule determines that the value of the evaluation result is 1 when the mean absolute error, root mean square error and mean relative error meet the preset thresholds at the same time, and establishes a precursor pathological model of vascular sclerosis based on the deep neural network learning algorithm; the abnormal prompt submodule is when the mean absolute error, root mean square error and mean relative error do not meet the preset thresholds, A value of 0 is determined as the evaluation result, indicating that the correlation between the historical analysis results and the real-time analysis results is too low, and an abnormal prompt is given. The above scheme analyzes and processes the analysis results through the model to obtain parameters with a high degree of correlation with vascular sclerosis, thereby ensuring the accuracy of the vascular sclerosis degree results, eliminating the tediousness of separately collecting heart rate, blood pressure and blood oxygen, shortening the examination time, and truly realizing the intelligent analysis of pulse wave signals. At the same time, the vascular sclerosis precursor pathology model compares the root mean square error and the mean relative error with the preset threshold value, and the evaluation result is determined by these two factors, which makes up for the inaccuracy of single factor measurement and further improves the accuracy of vascular sclerosis.

[0138] Example 11: Figure 11 As shown, based on Example 7, the real-time monitoring module provided by the embodiment of the present invention includes:

[0139] The state and vital signs data submodule is responsible for obtaining the timing acquisition frequency and collecting the state and vital signs data of the object to be detected according to the timing acquisition frequency;

[0140] The variance fluctuation value submodule is responsible for obtaining historical vital sign data and calculating the variance fluctuation value of state vital sign data and historical state data;

[0141] The judgment submodule is responsible for judging whether the health status of the subject to be tested is abnormal based on the pathological model of the precursor of vascular sclerosis and the variance fluctuation value, and generating a judgment result;

[0142] The abnormal result judgment submodule is responsible for triggering the pre-set alarm when the judgment result is abnormal, and sending a distress prompt to the terminal device pre-bound to the smart device sensor;

[0143] The normal result judgment submodule is responsible for storing the status vital sign data into the preset storage database when the judgment result is normal;

[0144] The working principle and beneficial effects of the above technical solution are as follows: the state vital sign data submodule of this embodiment obtains a timed collection frequency and collects state vital sign data of the subject to be detected based on the timed collection frequency; the variance fluctuation value submodule obtains historical vital sign data and calculates the variance fluctuation value of the state vital sign data and the historical state data; the judgment submodule determines whether the physical health status of the subject to be detected is abnormal based on the variance fluctuation value based on the pathological model of the precursor of vascular sclerosis and generates a judgment result; the abnormal result judgment submodule triggers a pre-set alarm when the judgment result is abnormal and sends a distress prompt to the terminal device pre-bound to the smart device sensor; the normal result judgment submodule stores the state vital sign data in a preset storage database when the judgment result is normal. The above solution uses the variance fluctuation value to determine whether the physical health status data of the subject to be detected is abnormal. The comparison of data can greatly improve the accuracy of the judgment and make it more precise. It eliminates the tedious traditional technology of doctors' judgment based on experience, saves medical resources, and improves the intelligence of data processing. The comparison of data also facilitates the intuitive understanding of the subject to be detected and improves the user experience of the subject to be detected.

[0145] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. A system for detecting vascular sclerosis parameters based on a light sensor, characterized in that: Include: The feature extraction module is responsible for acquiring and processing the original pulse wave signal through the preset smart device sensor and determining the pulse feature points; The data analysis module is responsible for transmitting the pulse feature points to the preset big data analysis platform for analysis and calculation, determining the vital sign data, and transmitting the vital sign data to the preset artificial intelligence algorithm for analysis and processing to determine the analysis results, which are used to determine whether the subject to be tested has signs of vascular sclerosis; Establish a pathological model module, which is responsible for establishing a pathological model of the precursors of vascular sclerosis by regularly collecting and analyzing results and transmitting them to a preset simulation device based on a deep neural network learning algorithm; The real-time monitoring module is responsible for determining whether the subject's vital signs are abnormal in real time based on the pathological model of vascular sclerosis precursors. If the vital signs are abnormal, it will issue an abnormal reminder and formulate corresponding treatment strategies in a timely manner. The process of transmitting vital sign data to the preset artificial intelligence algorithm for analysis and processing includes: Obtain pulse wave feature points and transmit them to a big data analysis platform; Based on the big data analysis platform, the incoming pulse wave signal characteristics are analyzed and calculated to obtain vital sign data; wherein the vital sign data includes at least heart rate data, blood pressure data and blood oxygen data; Through a preset artificial intelligence algorithm, all frequent data are first found from the physical sign data, and the support of the frequent data is greater than or equal to the minimum support threshold. Association rules are generated from the frequent data, and the confidence is calculated. The association rules with confidence greater than or equal to the minimum confidence threshold are retained. The physical sign data of the subject to be tested are mined and screened, and the corresponding treatment plan is launched and the analysis results are determined. Establish pathological model module, including: The regular collection submodule is responsible for regular collection and obtaining analysis results; The difference calculation submodule is responsible for calculating the predicted analysis results in the simulation device based on the analysis results and determining the difference between the analysis results and the predicted analysis results; The evaluation submodule is responsible for evaluating the difference and determining the evaluation results; A pathology model submodule is established, which is responsible for determining the evaluation result value to be 1 when the mean absolute error, root mean square error, and mean relative error all meet the preset thresholds. Based on the deep neural network learning algorithm, a pathology model for the precursor of vascular sclerosis is established; The abnormality prompt submodule is responsible for determining that the value of the evaluation result is 0 when the mean absolute error, root mean square error, and mean relative error do not meet the preset thresholds, indicating that the correlation between the historical analysis results and the real-time analysis results is too small, and an abnormality prompt is issued.

2. The system for detecting vascular sclerosis parameters based on a light sensor according to claim 1, wherein: Feature extraction module, including: The original pulse wave signal acquisition submodule is responsible for regularly detecting the pulse sign information of the subject to be detected through the preset wearable device sensor, and determining the original pulse wave signal of the subject to be detected based on the pulse sign information; The original pulse wave signal processing submodule is responsible for performing noise reduction, baseline removal and wavelet decomposition on the original pulse wave signal to determine the preprocessed pulse wave signal; The pulse wave feature point reconstruction submodule is responsible for collecting and reconstructing the pulse wave feature signal points based on the preprocessed pulse wave signal, and determining the pulse wave feature points.

3. The system for detecting vascular sclerosis parameters based on a light sensor according to claim 1, wherein: Data analysis module, including: The transmission submodule is responsible for obtaining pulse wave feature points and transmitting them to the big data analysis platform; The parsing submodule is responsible for analyzing and calculating the vital sign data by analyzing the incoming pulse wave signal characteristics based on the big data analysis platform. The vital sign data includes at least heart rate data, blood pressure data, and blood oxygen data. The analysis unit submodule is responsible for mining and screening the vital sign data of the test subject through a preset artificial intelligence algorithm, launching corresponding treatment plans, and determining the analysis results.

Images