COPD (chronic obstructive pulmonary disease) combined sarcopenia intelligent screening system and method based on big data

Abstract

The invention relates to the technical field of medical data processing, in particular to a COPD combined sarcopenia intelligent screening system and method based on big data, and the system comprises a diagnosis screening unit which calculates an observation judgment value of a COPD patient; preliminarily screening out high-risk patients, target patients and to-be-determined patients according to the observation judgment values; the interval analysis unit is used for acquiring a definite diagnosis time difference between the definite diagnosis of sarcopenia and the definite diagnosis of COPD, and constructing a function of the definite diagnosis time difference about the age and GOLD classification of the patient; determining a recheck interval based on a definite diagnosis time difference function; the tracking evaluation unit is used for generating a suggested reexamination signal based on the activity evaluation value of the target patient in combination with a reexamination interval; and the recheck analysis unit is used for analyzing and calculating a COPD induction index based on a tumor necrosis factor alpha index, an interleukin 6 index and a human serum albumin index of a rechecked patient, comparing and analyzing data sets detected twice, calculating a rescreening evaluation value by combining the COPD induction index, and marking a high-risk patient based on the rescreening evaluation value.

Description

Technical Field

[0001] The present invention relates to the technical field of medical data processing, and in particular to a big data-based intelligent screening system and method for COPD combined with sarcopenia. Background Art

[0002] There is a close, bidirectional interaction between COPD (chronic obstructive pulmonary disease) and sarcopenia, each reinforcing the other and significantly impacting patients' quality of life and prognosis. COPD and sarcopenia form a vicious cycle through inflammation, hypoxia, activity restriction, and metabolic imbalance. Early screening and multimodal interventions involving exercise, nutrition, and medication are crucial to break the cycle and improve prognosis.

[0003] The existing intelligent sarcopenia screening system usually uses the Asian Sarcopenia Diagnostic Standard (AWGS) for sarcopenia screening, but this diagnostic standard is not targeted, especially for COPD patients with limited lung function. Some tests in the diagnostic standard are difficult to meet the standards, increasing the probability of misdiagnosis. In addition, muscle loss is a progressive medical process, and the diagnosis in the existing technology is based on current muscle data, which is difficult to carry out continuous monitoring and diagnosis, easily increasing the probability of misdiagnosis and missed diagnosis, and has certain defects. Summary of the Invention

[0004] In response to the above-mentioned shortcomings of the existing technology, the present invention provides a big data-based intelligent screening system and method for COPD combined with sarcopenia, which can effectively solve the problem that the existing technology cannot track, monitor, diagnose and analyze COPD patients.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0006] The present invention provides a big data-based intelligent screening system and method for COPD combined with sarcopenia, which at least includes:

[0007] The diagnostic screening unit calculates the grip strength and physical fitness assessment values of COPD patients, and calculates the muscle mass assessment value based on the COPD patients' BMI index, body fat percentage, and upper arm circumference. The grip strength, muscle mass, and physical fitness assessment values are combined into a data set and combined with the COPD patients' lung function assessment values to calculate the observation judgment value;

[0008] Based on the observed judgment values, high-risk patients, target patients who need to be followed up and monitored, and pending patients who need regular reexamination are preliminarily screened;

[0009] Interval analysis unit: Build a database of COPD patients with combined sarcopenia, obtain the time difference between the diagnosis of sarcopenia and COPD, and construct a function of the time difference in diagnosis with respect to the patient's age and GOLD grade based on the database;

[0010] Determine the reexamination intervals for target patients and pending patients based on the time difference function for diagnosis;

[0011] The tracking and evaluation unit calculates the activity evaluation value based on the target patient's step data, exercise data, and sleep data, calculates the re-examination recommendation value based on the re-examination interval, and generates a re-examination recommendation signal based on the re-examination recommendation value;

[0012] The re-examination analysis unit calculates the COPD induction index based on the re-examination patient's tumor necrosis factor α, interleukin 6, and human serum albumin indicators, compares and analyzes the data sets of the two tests, calculates the re-screening assessment value based on the COPD induction index, and marks high-risk patients based on the re-screening assessment value;

[0013] Instrument retesting was performed on high-risk patients.

[0014] Furthermore, the process of obtaining the grip strength evaluation value is as follows:

[0015] COPD patients were tested for grip strength multiple times, and the maximum and minimum test results were recorded. The difference between the maximum and minimum values was calculated and divided by the number of grip strength tests to obtain the grip strength attenuation value. The maximum grip strength and grip strength attenuation values were divided by the preset gender grip strength baseline value and grip strength attenuation baseline value, respectively, and the sum was used to obtain the grip strength assessment value. The gender grip strength baseline value depends on the gender of the COPD patient.

[0016] Furthermore, the calculation formula for the muscle mass assessment value α2 is specifically as follows: in:

[0017] β AMA , β BFP , β BMI They represent the normalized upper limb muscle area AMA, body mass index, and body fat percentage, respectively; λ1 and λ2 are the preset weight coefficients; and λ3 is the preset gender determination coefficient, which depends on the gender of the COPD patient.

[0018] Furthermore, the retest interval determination process is as follows:

[0019] The medical database included patient age, COPD diagnosis time, GOLD grade, and sarcopenia diagnosis time. The time difference between the sarcopenia diagnosis time and the COPD diagnosis time was calculated to obtain the diagnosis time difference;

[0020] Construct a histogram of the time difference in diagnosis of all COPD patients with sarcopenia, which is recorded as the first histogram. The month corresponding to the highest column in the first histogram is used as the benchmark interval threshold T;

[0021] When the patient's age and GOLD grade are fixed values, the corresponding time difference of diagnosis is taken as the month corresponding to the highest column in the corresponding time difference bar chart. Let the patient's age and GOLD grade be x and y respectively, and construct the mathematical model of time difference of diagnosis. Where z represents the time difference of diagnosis, η1 and η2 are regression coefficients;

[0022] The data in the medical consultation database are used to train the mathematical model to estimate the regression coefficient and obtain the time difference calculation function f(x, y);

[0023] The patient age and GOLD grade of the pending patient are input to obtain the time difference for diagnosis, and the time difference for diagnosis is multiplied by the preset reduction coefficient to obtain the re-examination interval. The reduction coefficient is a constant less than 1 and greater than 0.

[0024] Furthermore, the activity evaluation value calculation process is as follows:

[0025] Obtain a line graph of the target patient's daily step count growth over time, obtain the target patient's daily exercise time interval as the exercise interval, obtain the step count growth value within the exercise interval, and subtract the sum of the step count growth values corresponding to all exercise intervals from the total daily step count to obtain the slow walking step count;

[0026] Calculate the target patient's average daily slow walking steps and average daily exercise duration, and normalize them with the target patient's average daily sleep duration to obtain dimensionless values, which are denoted as δ W , δ E , δ S , substitute into the formula EX=μ1*δ E +μ2*δ W -μ3*δ S The activity evaluation value EX is obtained by calculation, where μ1, μ2, and μ3 are all preset weight coefficients.

[0027] Furthermore, the process of determining the re-inspection recommendation value based on the activity evaluation value is as follows:

[0028] Obtain the target patient's activity evaluation value EX and reexamination interval, and substitute them into the formula Calculate and get the recommended value for re-inspection in:

[0029] EX all is the preset motion difference threshold, T RC Represents the retest interval, T now represents the monitoring time of the mobile monitoring device, EX′ represents the daily average activity threshold, and k′ is the preset adjustment coefficient.

[0030] Furthermore, the calculation process of the re-screening evaluation value is as follows:

[0031] Obtaining the grip strength assessment values, muscle mass assessment values, and physical fitness assessment values obtained during the primary screening process to form a primary data set; obtaining the grip strength assessment values, muscle mass assessment values, and physical fitness assessment values obtained during the re-examination process to form a re-examination data set; and calculating the grip strength difference value, muscle mass difference value, and physical fitness difference value based on the primary data set and the re-examination data set;

[0032] The grip strength difference, muscle mass difference, and physical fitness difference were normalized and taken as dimensionless values. Combined with the COPD induction index, they were multiplied by the corresponding preset weight coefficients and the sum was obtained to obtain the re-screening assessment value.

[0033] Furthermore, the COPD induction index calculation process is as follows:

[0034] Obtain the tumor necrosis factor α index and interleukin 6 index of the re-examination patient, record them as TNF and IL respectively, and preset the corresponding baseline values TNF′ and IL′ respectively. Substitute them into the formula The inflammatory impact index IS is calculated by Indicates that the value is not less than The smallest integer;

[0035] Obtain the human serum albumin index of the re-examination patient as ALB and substitute it into the formula The nutritional impact index NS is calculated by Indicates that the value is not less than ALB′ is the preset human serum albumin reference value, and R is the preset constant reference value;

[0036] The COPD induction index was obtained by multiplying the nutritional impact index by the preset proportional coefficient and then subtracting the inflammatory impact index.

[0037] A big data-based intelligent screening method for COPD combined with sarcopenia, applied to the above-mentioned big data-based intelligent screening system for COPD combined with sarcopenia, includes the following steps:

[0038] Step 1: Conduct grip strength and physical fitness tests on COPD patients. Calculate observation judgment values based on the test results, combined with the COPD patients' body mass index, body fat percentage, and upper arm circumference. Based on the observation judgment values, select high-risk patients who require instrument re-examination, COPD patients who require follow-up monitoring as target patients, and pending patients who require regular re-examination.

[0039] Step 2: A medical database was constructed based on the medical records of multiple patients with COPD and sarcopenia. The time difference between the diagnosis time of sarcopenia and the diagnosis time of COPD was calculated to obtain the diagnosis time difference. A relationship function between the diagnosis time difference and the patient's age and GOLD grade was constructed based on the medical database, denoted as the diagnosis time difference calculation function. The corresponding re-examination interval was calculated based on the patient's age, GOLD grade and the diagnosis time difference calculation function;

[0040] Step 3: Obtain the target patient's step count, exercise, and sleep data, analyze and calculate the activity assessment value, and combine the analysis and calculation of the target patient's corresponding re-examination interval to obtain a re-examination recommendation value. When the re-examination recommendation value is less than the re-examination recommendation threshold, a re-examination recommendation signal is generated;

[0041] Step 4: Obtain the re-examination data of the re-examination patient and compare it with the detection data in the primary screening process, analyze and calculate the re-screening evaluation value of the re-examination patient, and when the re-screening evaluation value is greater than the re-screening evaluation threshold, mark the re-examination patient as a high-risk patient and perform instrument re-examination on the high-risk patient.

[0042] Compared with the prior art, the technical solution provided by the present invention has the following beneficial effects:

[0043] 1. The total evaluation value of the present invention can comprehensively reflect the degree of match between the muscle performance of the initially screened patient and the sarcopenia performance from the three aspects of muscle mass, muscle strength and physical function. Compared with the AWGS diagnostic standard in the prior art, it can further quantify the patient's muscle performance rather than simply outputting the result based on threshold comparison. It is less dependent on medical equipment and can be measured and calculated without complex instruments, which is helpful for adaptive adjustments in subsequent processes. The observation judgment value obtained by correcting the total evaluation value in combination with the COPD evaluation value can minimize the impact of the COPD patient's lung function on the muscle performance evaluation, thereby obtaining an observation judgment value that is more in line with the actual physical condition of the initially screened patient.

[0044] 2. The present invention can recommend target patients to undergo re-examination at an appropriate time based on their daily exercise conditions, thereby diagnosing sarcopenia in the target patients as timely as possible and reducing the damage to the target patients' bodies caused by delayed diagnosis of sarcopenia. In addition, by calculating the re-screening evaluation value, not only can the sarcopenia of the re-examined patients be evaluated from the perspective of muscle function decline, but the results can also be corrected from the perspective of the biological mechanism of COPD-induced sarcopenia, thereby screening out patients who are prone to COPD combined with sarcopenia from two aspects: external functional performance and changes in internal biological indicators, thereby improving the screening efficiency of the screening system and the detection efficiency of the detection instrument. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. Those skilled in the art can also derive other drawings based on these drawings without inventive effort.

[0046] Figure 1 It is the overall module block diagram of the present invention. DETAILED DESCRIPTION

[0047] To make the purpose, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0048] The present invention will be further described below with reference to the embodiments.

[0049] See Figure 1 , a big data-based intelligent screening system for COPD combined with sarcopenia, including at least:

[0050] The diagnostic screening unit conducts primary screening for confirmed COPD patients. The primary screening includes static assessments from multiple angles. Based on the analysis of the static assessment results, observation judgment values are obtained. Based on the observation judgment values, COPD patients who need to be tracked and monitored are selected as target patients, including:

[0051] The COPD patients who have been diagnosed are recorded as the initial screening patients. The initial screening patients are tested for grip strength multiple times. The maximum and minimum values of the test results are recorded. The difference between the maximum and minimum values is calculated and divided by the number of grip strength tests to obtain the grip strength attenuation value. Substitute it into the formula The grip strength evaluation value α1 is calculated in , where Gs max 、 They represent the maximum grip strength and grip strength attenuation value respectively, Gs′ is the preset gender grip strength reference value. When the initial screening patient is female, the gender grip strength threshold is 18, and when the initial screening patient is male, the gender determination coefficient is 28. is the preset grip strength attenuation benchmark value;

[0052] Obtain the body mass index (BMI, calculated by dividing weight by the square of height), body fat percentage, and upper arm circumference of the initial screening patients. Calculate the upper limb muscle area (AMA) of the initial screening patients based on the upper arm circumference. The calculation formula is upper limb muscle area (AMA) = [upper arm circumference - triceps skin fold thickness × 3.14] 2 / 4π - gender influence coefficient. When the initial screening patient is female, the gender influence coefficient is 6.5. When the initial screening patient is male, the gender influence coefficient is 10. The upper limb muscle area (AMA) is combined with the body mass index and body fat percentage to obtain the muscle mass assessment value. The specific calculation formula is: Among them, α2 represents the muscle mass assessment value, β AMA , β BFP , β BMI They represent the normalized upper limb muscle area AMA, body mass index, and body fat percentage (dimensionless values), respectively; λ1 and λ2 are preset weight coefficients; λ3 is the preset gender determination coefficient. When the initial screening patient is female, the gender determination coefficient is 0.25; when the initial screening patient is male, the gender determination coefficient is 0.20;

[0053] It should be noted that the muscle mass assessment value obtained by comprehensively calculating the influencing values of upper limb muscle area, body mass index, and body fat percentage can be used to evaluate the patient's muscle content. The larger the muscle mass assessment value usually means the higher the patient's muscle content. The muscle assessment value can be used to screen out patients with abnormal muscle content. Compared with the BIA and DXA methods in the existing technology, this assessment method is less dependent on medical equipment, and can be measured and calculated without complex instruments. It is more suitable for application in primary medical institutions to screen for sarcopenia.

[0054] Perform a gait speed test on the initial screening patients and record the patient's walking speed. If the gait speed test is not available, perform a sit-up test on the initial screening patients and record the time taken to sit up for n preset sit-ups. Based on the gait speed test results or the sit-up test results of the initial screening patients, the physical function of the initial screening patients is evaluated to obtain a physical fitness assessment value. When the initial screening patients undergo a gait speed test, the calculation formula for the physical fitness assessment value is α3=exp(λ4*(v1-1)). When the initial screening patients undergo a sit-up test, the calculation formula for the physical fitness assessment value is Where exp(x) represents an exponential function with the natural constant e as the base and the exponent x as the exponent, λ4 is the preset weight coefficient, v1 and t1 represent the walking speed and the time it takes to get up and sit down, respectively, of the patients undergoing initial screening.

[0055] It should be noted that the gait speed test and sit-up test are existing technologies and are physical function testing methods in the Asian Sarcopenia Diagnostic Criteria (AWGS), so they will not be discussed in detail here. The physical fitness assessment value reflects the physical fitness level of the initial screening patient. When the physical fitness assessment value is higher, it means that the physical fitness level of the initial screening patient is higher and the likelihood of sarcopenia is lower.

[0056] The lung function of the initially screened patients was rated based on the GOLD grading standard and divided into GOLD grade 1, GOLD grade 2, GOLD grade 3 and GOLD grade 4. A lung function assessment value α4 was assigned to the initially screened patients based on the GOLD grade. The lung function assessment values corresponding to GOLD grade 1, GOLD grade 2, GOLD grade 3 and GOLD grade 4 were 1, 2, 3 and 4 respectively.

[0057] The grip strength evaluation value α1, muscle mass evaluation value α2, and physical fitness evaluation value α3 are normalized and summed to obtain the total evaluation value α all , combined with the lung function assessment value α4, the total assessment value calculation formula is modified to obtain the observation judgment value β. The modified formula is β=α all -k RE *α4, where k RE is the preset correction factor.

[0058] It should be noted that the total assessment value can comprehensively reflect the degree of fit between the muscle performance of the initially screened patients and the sarcopenia performance from three aspects: muscle mass (magnitude), muscle strength and physical function. Compared with the AWGS diagnostic standard in the existing technology, it can further quantify the patient's muscle performance (applicable to patients of different degrees), rather than simply comparing the threshold to output the result. It can make more detailed result outputs based on the patient's muscle performance, which is helpful for adaptive adjustments in the subsequent process. In addition, the observation judgment value obtained by correcting the total assessment value in combination with the COPD assessment value can minimize the impact of the lung function of COPD patients on the muscle performance assessment, thereby obtaining an observation judgment value that is more in line with the actual physical condition of the initially screened patients.

[0059] It is worth noting that the decreased lung function of COPD patients will cause them to not show their actual muscle level during grip strength tests and physical fitness tests (i.e., walking speed tests or sit-up tests). The specific mechanism of action is: COPD patients are unable to hold their breath and exert force fully due to rapid breathing during grip strength tests, resulting in low grip strength values (10%-15% lower than the actual value). Limited lung function makes it difficult for COPD patients to exhale steadily, thereby limiting their walking speed and sit-up speed.

[0060] A first observation threshold and a second observation threshold are preset, and the first observation threshold is greater than the second observation threshold. When the observation judgment value is greater than the first observation threshold, the initially screened patient is recorded as a high-risk patient, and the high-risk patient is re-examined with an instrument. When the observation judgment value is less than or equal to the first observation threshold and greater than the second observation threshold, the initially screened patient is marked as a target patient, and a mobile monitoring device is used to monitor the patient's daily activity data. When the observation judgment value is less than or equal to the second observation threshold, the initially screened patient is marked as a pending patient, and the pending patient is re-examined regularly. The re-examination interval of the pending patient is obtained based on big data analysis, that is, the pending patient is re-examined after the preset re-examination interval.

[0061] It should be noted that high-risk patients refer to COPD patients whose current characteristics are consistent with sarcopenia (for example, patients with very low muscle dimensions and severe decline in muscle capacity), target patients refer to COPD patients whose characteristics are consistent with sarcopenia (for example, patients with small muscle dimensions but normal grip strength and physical function), and pending patients refer to COPD patients with very few or no characteristics consistent with sarcopenia (for example, patients with normal muscle function).

[0062] The interval analysis unit analyzes and calculates the retest interval. The calculation process is as follows:

[0063] Obtain medical records of multiple patients with COPD and sarcopenia, extract the patient's age (age at the time of COPD diagnosis), COPD diagnosis time, GOLD grade and sarcopenia diagnosis time from the medical records to form a medical database, calculate the time difference between the sarcopenia diagnosis time and the COPD diagnosis time to obtain the diagnosis time difference, and construct a mathematical model of the diagnosis time difference with the patient's age and GOLD grade (both dimensionless values) as independent variables. In the mathematical model, the diagnosis time difference is the dependent variable, and the mathematical model is trained based on the data in the medical database to obtain a diagnosis time difference calculation function f(x, y), where x and y represent the dimensionless values corresponding to the patient's age and GOLD grade, respectively. The patient's age and GOLD grade of the patient to be determined are input to obtain the diagnosis time difference, and the diagnosis time difference is multiplied by a preset reduction coefficient to obtain the re-examination interval, where the reduction coefficient is a constant less than 1 and greater than 0 (in a specific embodiment, the value is 0.5).

[0064] It should be noted that the actual onset time of sarcopenia is difficult to determine in the prior art. Therefore, the diagnosis time of sarcopenia is used as the analysis object in this application, because there is a clear correlation between the diagnosis time of sarcopenia and the actual onset time of sarcopenia. An important indicator of sarcopenia is the impact on physical function, and this indicator is more easily perceived in the patient's daily life, thereby inducing the patient to diagnose sarcopenia. Therefore, the diagnosis time difference is used as the

[0065] Specifically, a histogram of the time difference in diagnosis of all COPD patients with sarcopenia is constructed and recorded as the first histogram (the unit length of the horizontal axis in the histogram is month, and each single column represents the time difference in diagnosis data within the corresponding month. For example, the number of time differences in diagnosis corresponding to December in the histogram is 266, which means that 266 COPD patients were diagnosed with sarcopenia within the 12th month after the diagnosis of COPD). The month corresponding to the highest column in the first histogram is taken as the reference interval threshold T. The patient age and GOLD grade are set to x and y respectively. When the patient age and GOLD grade are fixed values, the corresponding time difference in diagnosis is taken as the month corresponding to the highest column in the corresponding time difference in diagnosis histogram, and a mathematical model of the time difference in diagnosis is constructed. Where z represents the time difference to diagnosis, η1 and η2 are regression coefficients, and the data in the medical database are used to train the mathematical model to estimate the regression coefficients and obtain the time difference to diagnosis calculation function f(x, y).

[0066] In a specific embodiment, the T value obtained based on the database is 12, and η1 and η2 are 0.05 and 0.3 respectively.

[0067] Older patients or those with higher GOLD grades may need shorter reexamination intervals because they may lose muscle faster. Compared with the fixed reexamination intervals in existing technologies, the reexamination intervals determined based on the patient's age and GOLD grade can be adaptively optimized and adjusted according to the patient's actual physical conditions. In particular, the reexamination intervals can be shortened for high-risk groups (older people and those with higher GOLD grades), thereby improving the detection rate and prevention effect of sarcopenia. For low-risk patients, the reexamination intervals will not be shortened excessively, thereby reducing the impact of frequent reexaminations on the lives and psychological pressure of low-risk patients.

[0068] The tracking and evaluation unit obtains the re-examination interval corresponding to the target patient, analyzes and calculates the activity evaluation value of the target patient based on multiple data of the target patient recorded by the mobile monitoring device, and obtains the re-examination recommendation value based on the activity evaluation value combined with the re-examination interval analysis. When the re-examination recommendation value is less than the re-examination recommendation threshold (in a specific embodiment, the value is 0), a re-examination recommendation signal is generated, and a "sarcopenia diagnosis is recommended in the near future" instruction is sent to the target patient through the mobile monitoring device to remind the target patient to undergo a re-examination.

[0069] Specifically, the activity evaluation value calculation process is as follows:

[0070] Obtain a line graph of the target patient's daily step count growth over time, obtain the target patient's daily exercise time interval and record it as the exercise interval (i.e., the time interval in the exercise state), obtain the step count growth value within the exercise interval, and subtract the sum of the step count growth values corresponding to all exercise intervals from the total daily step count to obtain the slow walking step count (by eliminating the steps within the exercise interval from the total step count, it is possible to distinguish the step count growth caused by running exercise and daily slow walking, so that running exercise is only used as the calculation of the activity evaluation value of exercise behavior, avoiding repeated calculations that affect the accuracy of the final result). Calculate the target patient's average daily slow walking step count and average daily exercise time, and combine them with the target patient's average daily sleep time for normalization to obtain dimensionless values, which are recorded as δ W , δ E , δ S (corresponding to the average daily number of slow steps, average daily exercise time, and average daily sleep time, respectively), substitute into the formula EX = μ1*δ E +μ2*δ W -μ3*δ S The activity evaluation value EX is obtained by calculation, where μ1, μ2, and μ3 are all preset weight coefficients.

[0071] It should be noted that mobile monitoring devices can be smart bracelets and watches provided by medical institutions to patients, or they can be the patients' own smart bracelets and smart watches. Mobile monitoring devices can monitor patients' daily exercise data and transmit it to the sports app on the patient's mobile phone via Bluetooth, and then the sports app transmits it to the cloud through the communication network for processing and analysis. Daily exercise data includes exercise duration (the data recorded by the three-axis accelerometer and heart rate sensor inside the mobile monitoring device can determine whether the patient is in an exercise state), number of steps, and sleep duration, etc.

[0072] Furthermore, the process of determining the re-inspection recommendation value based on the activity evaluation value is as follows:

[0073] Obtain the target patient's activity evaluation value EX and reexamination interval, and substitute them into the formula Calculate and get the recommended value for re-inspection Among them EX all is the preset motion difference threshold, T RC Represents the retest interval, T now represents the monitoring time of the mobile monitoring device, EX′ represents the daily average activity threshold, and k′ is the preset adjustment coefficient.

[0074] By calculating the recommended re-examination value, the target patient can be recommended for re-examination at the appropriate time based on the target patient's daily exercise situation, so that the target patient can be diagnosed with sarcopenia as timely as possible, reducing the damage caused to the target patient's body by the delay in diagnosis of sarcopenia. This is because the physiological manifestations of sarcopenia in COPD patients are often a progressive process and are closely related to the reduced amount of exercise in COPD patients. When COPD patients lack exercise after being diagnosed with COPD, the risk of developing sarcopenia increases. Therefore, more frequent re-examinations are needed for COPD patients who lack exercise to diagnose sarcopenia in a timely manner.

[0075] The re-examination analysis unit obtains the re-examination data of the re-examination patient and compares it with the test data in the primary screening process, analyzes and calculates the re-screening evaluation value of the re-examination patient. When the re-screening evaluation value is greater than the re-screening evaluation threshold, the re-examination patient is marked as a high-risk patient, and the high-risk patient is re-examined by the instrument.

[0076] Specifically, the calculation process of the re-screening evaluation value is as follows:

[0077] The grip strength assessment values, muscle mass assessment values, and physical fitness assessment values obtained during the primary screening process are obtained to form a primary data set. The grip strength assessment values, muscle mass assessment values, and physical fitness assessment values obtained during the re-examination process are obtained to form a re-examination data set. Based on the primary data set and the re-examination data set, the grip strength difference value Δα1, the muscle mass difference value Δα2, and the physical fitness difference value Δα3 are calculated;

[0078] Obtain the tumor necrosis factor α (TNF-α) and interleukin 6 (IL-6) indicators of the re-examination patients, record them as TNF and IL respectively, and preset the corresponding baseline values TNF′ and IL′ respectively. Substitute them into the formula The inflammatory impact index IS is calculated by Indicates that the value is not less than The smallest integer of the human albumin index of the re-examination patient is obtained as ALB, and it is substituted into the formula The nutritional impact index NS is calculated by Indicates that the value is not less than ALB′ is a preset human serum albumin reference value, R is a preset constant reference value (5 in a specific embodiment), and the COPD induction index is obtained by multiplying the nutritional impact index by the preset proportional coefficient and then subtracting the inflammation impact index;

[0079] It should be noted that the human serum albumin index labeled ALB, the tumor necrosis factor α index TNF, and the interleukin 6 index IL are all dimensionless values. The mechanism by which COPD causes sarcopenia is mainly that the airways and lung tissues of COPD patients continuously release proinflammatory factors such as TNF-α, IL-6, and CRP, which activate the ubiquitin-proteasome system (UPS) and accelerate the breakdown of muscle protein. In addition, the reduction of human serum albumin will affect the patient's body recovery. In other words, the lower the human serum albumin index ALB, the higher the tumor necrosis factor α index TNF and the interleukin 6 index IL, the more likely it is that COPD patients will suffer from sarcopenia. Therefore, the COPD induction index can reflect the degree of influence of COPD symptom characteristic indicators on patients suffering from sarcopenia. Generally speaking, the larger the COPD assessment value, the greater the probability that the patient will suffer from sarcopenia, and it needs to be monitored closely.

[0080] The grip strength difference, muscle mass difference, and physical fitness difference were normalized and taken as dimensionless values. Combined with the COPD induction index, they were multiplied by the corresponding preset weight coefficients and the sum was obtained to obtain the re-screening assessment value.

[0081] It should be noted that the re-screening evaluation value can not only assess the prevalence of sarcopenia in re-examined patients from the perspective of muscle function decline, but also correct the results from the perspective of the biological mechanism of COPD-induced sarcopenia, thereby screening out patients who are prone to COPD combined with sarcopenia from both external functional performance and changes in internal biological indicators, thereby improving the screening efficiency of the screening system and the detection efficiency of the detection instrument.

[0082] A big data-based intelligent screening method for COPD combined with sarcopenia, comprising the following steps:

[0083] Step 1: Conduct grip strength and physical fitness tests on confirmed COPD patients. Calculate observation judgment values based on the test results combined with the COPD patients' body mass index, body fat percentage, and upper arm circumference. Based on the observation judgment values, select high-risk patients who require instrument re-examination, COPD patients who require follow-up monitoring as target patients, and pending patients who require regular re-examination.

[0084] By comprehensively calculating observation values through grip strength tests, physical fitness tests, BMI, body fat percentage, upper arm circumference and other indicators, patients are divided into three categories: high-risk, to be followed (target patients) and pending, to achieve tiered management, avoid a one-size-fits-all screening approach, prioritize high-risk patients, optimize the allocation of medical resources, improve applicability, and make it applicable to community medical institutions.

[0085] Step 2: A medical consultation database is constructed based on the medical records of multiple patients with COPD and sarcopenia. The medical consultation database includes the patient's age, COPD diagnosis time, GOLD grade, and sarcopenia diagnosis time. The time difference between the sarcopenia diagnosis time and the COPD diagnosis time is calculated to obtain the diagnosis time difference. A relationship function between the diagnosis time difference and the patient's age and GOLD grade is constructed based on the medical consultation database, which is recorded as the diagnosis time difference calculation function. The corresponding re-examination interval is calculated based on the patient's age, GOLD grade, and the diagnosis time difference calculation function;

[0086] Based on the historical medical database, a relationship function between the time difference in diagnosis and age and GOLD grade is constructed, and the re-examination interval is dynamically calculated, taking into account the severity of the disease (GOLD grade) and the patient's age (the risk of sarcopenia increases with age). This makes the re-examination interval personalized and more in line with the patient's actual risk, avoiding excessive examination or missed examination caused by fixed re-examination cycles.

[0087] Step 3: Obtain the target patient's step count data, exercise data, and sleep data, analyze and calculate the target patient's activity assessment value, and analyze and calculate the re-examination recommendation value based on the activity assessment value combined with the target patient's corresponding re-examination interval. When the re-examination recommendation value is less than the re-examination recommendation threshold, generate a re-examination recommendation signal;

[0088] Activity assessment values are calculated through daily activity data such as steps, exercise, and sleep, and re-examination recommendation values are dynamically adjusted. Static clinical data (such as GOLD grade) is combined with dynamic behavioral data to more sensitively reflect changes in patients' functional status. When activity ability decreases (which may indicate the progression of sarcopenia), re-examination recommendations are generated in a timely manner to achieve early intervention.

[0089] Step 4: Obtain the re-examination data of the re-examination patient and compare it with the detection data in the primary screening process, analyze and calculate the re-screening evaluation value of the re-examination patient, and when the re-screening evaluation value is greater than the re-screening evaluation threshold, mark the re-examination patient as a high-risk patient and perform instrument re-examination on the high-risk patient.

[0090] The above method can monitor sarcopenia in COPD patients, realize multimodal data integration, dynamic risk stratification and closed-loop management, thereby achieving accurate screening and monitoring of COPD combined with sarcopenia, which is both scientific, practical and scalable.

[0091] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the protection scope of the technical solutions of the various embodiments of the present invention.

Claims

1. An intelligent screening system for COPD combined with sarcopenia based on big data, characterized by: include: The diagnostic screening unit calculates the grip strength and physical fitness assessment values of COPD patients, and calculates the muscle mass assessment value based on the COPD patients' BMI index, body fat percentage, and upper arm circumference. The grip strength, muscle mass, and physical fitness assessment values are combined into a data set and combined with the COPD patients' lung function assessment values to calculate the observation judgment value; Based on the observed judgment values, high-risk patients, target patients who need to be followed up and monitored, and pending patients who need regular reexamination are preliminarily screened; Interval analysis unit: Build a database of COPD patients with combined sarcopenia, obtain the time difference between the diagnosis of sarcopenia and COPD, and construct a function of the time difference in diagnosis with respect to the patient's age and GOLD grade based on the database; Determine the reexamination intervals for target patients and pending patients based on the time difference function for diagnosis; The tracking and evaluation unit calculates the activity evaluation value based on the target patient's step data, exercise data, and sleep data, calculates the re-examination recommendation value based on the re-examination interval, and generates a re-examination recommendation signal based on the re-examination recommendation value; The re-examination analysis unit calculates the COPD induction index based on the re-examination patient's tumor necrosis factor α, interleukin 6, and human serum albumin indicators, compares and analyzes the data sets of the two tests, calculates the re-screening assessment value based on the COPD induction index, and marks high-risk patients based on the re-screening assessment value; Instrument retesting was performed on high-risk patients.

2. The intelligent screening system for COPD combined with sarcopenia based on big data according to claim 1, characterized in that: The process of obtaining the grip strength assessment value is as follows: COPD patients were tested for grip strength multiple times, and the maximum and minimum test results were recorded. The difference between the maximum and minimum values was calculated and divided by the number of grip strength tests to obtain the grip strength attenuation value. The maximum grip strength and grip strength attenuation values were divided by the preset gender grip strength baseline value and grip strength attenuation baseline value, respectively, and the sum was used to obtain the grip strength assessment value. The gender grip strength baseline value depends on the gender of the COPD patient.

3. The intelligent screening system for COPD combined with sarcopenia based on big data according to claim 1, characterized in that: The calculation formula for muscle mass assessment value α2 is as follows: in: β AMA , β BFP , β BMI They represent the normalized upper limb muscle area AMA, body mass index, and body fat percentage, respectively; λ1 and λ2 are the preset weight coefficients; and λ3 is the preset gender determination coefficient, which depends on the gender of the COPD patient.

4. The intelligent screening system for COPD combined with sarcopenia based on big data according to claim 1, characterized in that: The retest interval determination process is as follows: The medical database included patient age, COPD diagnosis time, GOLD grade, and sarcopenia diagnosis time. The time difference between the sarcopenia diagnosis time and the COPD diagnosis time was calculated to obtain the diagnosis time difference; Construct a histogram of the time difference in diagnosis of all COPD patients with sarcopenia, which is recorded as the first histogram. The month corresponding to the highest column in the first histogram is used as the benchmark interval threshold T; When the patient's age and GOLD grade are fixed values, the corresponding time difference of diagnosis is taken as the month corresponding to the highest column in the corresponding time difference bar chart. Let the patient's age and GOLD grade be x and y respectively, and construct the mathematical model of time difference of diagnosis. Where z represents the time difference of diagnosis, η1 and η2 are regression coefficients; The data in the medical consultation database are used to train the mathematical model to estimate the regression coefficient and obtain the time difference calculation function f(x, y); The patient age and GOLD grade of the pending patient are input to obtain the time difference for diagnosis, and the time difference for diagnosis is multiplied by the preset reduction coefficient to obtain the re-examination interval. The reduction coefficient is a constant less than 1 and greater than 0.

5. The intelligent screening system for COPD combined with sarcopenia based on big data according to claim 1, characterized in that: The activity evaluation value calculation process is as follows: Obtain a line graph of the target patient's daily step count growth over time, obtain the target patient's daily exercise time interval as the exercise interval, obtain the step count growth value within the exercise interval, and subtract the sum of the step count growth values corresponding to all exercise intervals from the total daily step count to obtain the slow walking step count; Calculate the target patient's average daily slow walking steps and average daily exercise duration, and normalize them with the target patient's average daily sleep duration to obtain dimensionless values, which are denoted as δ W , δ E , δ S , substitute into the formula EX=μ1*δ E +μ2*δ W -μ3*δ S The activity evaluation value EX is obtained by calculation, where μ1, μ2, and μ3 are all preset weight coefficients.

6. The intelligent screening system for COPD combined with sarcopenia based on big data according to claim 5, characterized in that: The process of determining the re-inspection recommendation value based on the activity evaluation value is as follows: Obtain the target patient's activity evaluation value EX and reexamination interval, and substitute them into the formula Calculate and get the recommended value for re-inspection in: EX all is the preset motion difference threshold, T RC Represents the retest interval, T now represents the monitoring time of the mobile monitoring device, EX′ represents the daily average activity threshold, and k′ is the preset adjustment coefficient.

7. The intelligent screening system for COPD combined with sarcopenia based on big data according to claim 1, characterized in that: The calculation process of the re-screening evaluation value is as follows: Obtaining the grip strength assessment values, muscle mass assessment values, and physical fitness assessment values obtained during the primary screening process to form a primary data set; obtaining the grip strength assessment values, muscle mass assessment values, and physical fitness assessment values obtained during the re-examination process to form a re-examination data set; and calculating the grip strength difference value, muscle mass difference value, and physical fitness difference value based on the primary data set and the re-examination data set; The grip strength difference, muscle mass difference, and physical fitness difference were normalized and taken as dimensionless values. Combined with the COPD induction index, they were multiplied by the corresponding preset weight coefficients and the sum was obtained to obtain the re-screening assessment value.

8. The intelligent screening system for COPD combined with sarcopenia based on big data according to claim 7, characterized in that: The COPD inducibility index is calculated as follows: Obtain the tumor necrosis factor α index and interleukin 6 index of the re-examination patient, record them as TNF and IL respectively, and preset the corresponding baseline values TNF′ and IL′ respectively. Substitute them into the formula The inflammatory impact index IS is calculated by Indicates that the value is not less than The smallest integer; Obtain the human serum albumin index of the re-examination patient as ALB and substitute it into the formula The nutritional impact index NS is calculated by Indicates that the value is not less than ALB′ is the preset human serum albumin reference value, and R is the preset constant reference value; The COPD induction index was obtained by multiplying the nutritional impact index by the preset proportional coefficient and then subtracting the inflammatory impact index.

9. A big data-based intelligent screening method for COPD combined with sarcopenia, applied to a big data-based intelligent screening system for COPD combined with sarcopenia according to any one of claims 1 to 8, characterized in that: The following steps are involved: Step 1: Conduct grip strength and physical fitness tests on COPD patients. Calculate observation judgment values based on the test results, combined with the COPD patients' body mass index, body fat percentage, and upper arm circumference. Based on the observation judgment values, select high-risk patients who require instrument re-examination, COPD patients who require follow-up monitoring as target patients, and pending patients who require regular re-examination. Step 2: A medical database was constructed based on the medical records of multiple patients with COPD and sarcopenia. The time difference between the diagnosis time of sarcopenia and the diagnosis time of COPD was calculated to obtain the diagnosis time difference. A relationship function between the diagnosis time difference and the patient's age and GOLD grade was constructed based on the medical database, denoted as the diagnosis time difference calculation function. The corresponding re-examination interval was calculated based on the patient's age, GOLD grade and the diagnosis time difference calculation function; Step 3: Obtain the target patient's step count, exercise, and sleep data, analyze and calculate the activity assessment value, and combine the analysis and calculation of the target patient's corresponding re-examination interval to obtain a re-examination recommendation value. When the re-examination recommendation value is less than the re-examination recommendation threshold, a re-examination recommendation signal is generated; Step 4: Obtain the re-examination data of the re-examination patient and compare it with the detection data in the primary screening process, analyze and calculate the re-screening evaluation value of the re-examination patient, and when the re-screening evaluation value is greater than the re-screening evaluation threshold, mark the re-examination patient as a high-risk patient and perform instrument re-examination on the high-risk patient.

Images