Machine learning driven colorectal cancer kit detection data intelligent processing method

Abstract

The invention provides a machine learning driven colorectal cancer kit detection data intelligent processing method, and relates to the technical field of data processing, the method comprises the following steps: obtaining multiple groups of healthy sample detection data obtained by a colorectal cancer kit; determining a steady state center representing a health state based on each group of health sample detection data, and establishing a Jacobian matrix describing the health state through a regression fitting process; constructing a Lyapunov equation by taking the Jacobian matrix as input, and solving a positive definite matrix of the Lyapunov equation; establishing a Lyapunov energy function based on the positive definite matrix so as to quantify the detection data into energy cost deviating from a steady-state center; acquiring to-be-detected data of the colorectal cancer kit; inputting the data to be detected into the Lyapunov energy function, and solving the change rate of the Lyapunov energy function; and when the change rate of the Lyapunov energy function is less than zero, outputting that no colorectal cancer risk exists, otherwise, outputting that the colorectal cancer risk exists.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, in particular to a machine learning driven intelligent processing method for colorectal cancer kit detection data. BACKGROUND

[0002] The colorectal cancer kit is a diagnostic tool for detecting colorectal cancer-related biomarkers, including various reagents and detection methods, which can help early detection of colorectal cancer or predict its risk by analyzing specific molecular or genetic information in patient samples. These kits can be detected through blood, stool, tissue and other samples, and have the characteristics of rapid, non-invasive and high sensitivity.

[0003] It is necessary to perform intelligent processing of colorectal cancer kit detection data, because traditional data processing methods are often inefficient and susceptible to human interference, leading to misdiagnosis or missed diagnosis. Machine learning technology can improve the accuracy and reliability of data processing through large-scale data learning and analysis, identify potential abnormal patterns and minor changes, and provide precise diagnostic support for doctors, helping to achieve early detection and personalized treatment plan development, greatly improving clinical treatment effect and patient survival rate.

[0004] However, the existing intelligent processing of colorectal cancer kit detection data is usually based on experience and artificially defined classification threshold to determine whether there is a risk of colorectal cancer, which is difficult to achieve threshold-free classification, resulting in subjective interference in the processing result and insufficient accuracy. SUMMARY

[0005] In view of the above deficiencies of the prior art, the purpose of the embodiments of the present application is to provide a machine learning driven intelligent processing method for colorectal cancer kit detection data, which can solve the technical problems of the prior art that the existing intelligent processing of colorectal cancer kit detection data is usually based on experience and artificially defined classification threshold to determine whether there is a risk of colorectal cancer, which is difficult to achieve threshold-free classification, resulting in subjective interference in the processing result and insufficient accuracy.

[0006] The first aspect of the embodiments of the present application proposes a machine learning driven intelligent processing method for colorectal cancer kit detection data, comprising:

[0007] S1: obtaining a plurality of sets of health sample detection data obtained by a colorectal cancer kit;

[0008] S2: determining a steady state center representing a healthy state based on each set of health sample detection data, and establishing a Jacobian matrix describing the healthy state through a regression fitting process;

[0009] S3: constructing a Lyapunov equation with the Jacobian matrix as input, and solving a positive definite matrix of the Lyapunov equation;

[0010] S4: Establish a Lyapunov energy function based on a positive definite matrix to quantify the detection data as the energy cost of deviating from the steady-state center;

[0011] S5: Obtain the test data for the colorectal cancer kit;

[0012] S6: Input the data to be detected into the Lyapunov energy function and solve for the rate of change of the Lyapunov energy function;

[0013] S7: If the rate of change of the Lyapunov energy function is less than zero, the output indicates no risk of colorectal cancer; otherwise, the output indicates a risk of colorectal cancer.

[0014] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following:

[0015] In this embodiment of the invention, a steady-state center of health is determined by regression fitting of healthy sample data, and a Jacobian matrix is ​​established to describe the changing characteristics of health status. Then, a Lyapunov equation is constructed based on the Jacobian matrix and solved to obtain a positive definite matrix, thereby establishing a Lyapunov energy function to quantify the degree to which the detection data deviates from the healthy steady-state center. After the data to be detected is input into the Lyapunov energy function, the calculated rate of change reflects the degree of deviation of the data relative to the healthy steady-state center. If the rate of change is less than zero, it indicates that the deviation between the data to be detected and the healthy state is small, the data is in a stable state, and no abnormalities have occurred, therefore there is no risk of colorectal cancer. Conversely, if the rate of change is greater than zero, it indicates that the data to be detected deviates from the healthy steady-state center, and there is a large change or abnormality, suggesting a possible risk of colorectal cancer. This process, by accurately quantifying data deviation, avoids subjective interference from manually set thresholds, making risk assessment more objective and accurate. It improves the objectivity and accuracy of the detection results, avoids the setting of manually set thresholds, and can more accurately identify potential risks. Attached Figure Description

[0016] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Throughout the drawings, the same reference numerals denote the same parts. Obviously, the drawings described below are merely some embodiments of the present invention, and those skilled in the art can obtain other drawings based on these drawings without any creative effort.

[0017] Figure 1 This is a flowchart illustrating a machine learning-driven intelligent processing method for colorectal cancer reagent kit detection data provided in an embodiment of the present invention. Detailed Implementation

[0018] To enable those skilled in the art to better understand the technical solutions in the embodiments of the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. It should be understood that these descriptions are merely exemplary and are not intended to limit the scope of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0019] The following description, in conjunction with the accompanying drawings, details the machine learning-driven intelligent data processing method for colorectal cancer reagent kits provided in this invention through specific embodiments and application scenarios.

[0020] Reference manual attached Figure 1 The diagram illustrates a flowchart of a machine learning-driven intelligent processing method for colorectal cancer reagent kit detection data provided in an embodiment of the present invention.

[0021] This invention provides a machine learning-driven intelligent processing method for colorectal cancer reagent kit detection data, which may include the following steps:

[0022] S1: Obtain test data from multiple healthy samples obtained using the colorectal cancer kit.

[0023] The health sample testing data comes from the historical test results of normal healthy individuals, serving as the basis for building the health status model. By collecting data from different health samples, sufficient samples are provided for homeostasis centers and health status modeling, so as to more accurately describe the difference between health status and cancer risk.

[0024] In one possible implementation, the health sample testing data includes multiple colorectal cancer biomarker data, which include tumor biomarker data, gene mutation data, DNA methylation data, circulating tumor DNA data, and microbiome data.

[0025] In one possible implementation, tumor marker data includes carcinoembryonic antigen (CEA) data, CA19-9 data, M2-PK data, and CA72-4 data. Gene mutation data includes KRAS gene mutation data, BRAF gene mutation data, TP53 gene mutation data, and PIK3CA gene mutation data. DNA methylation data includes MLH1 gene methylation data, CDKN2A gene methylation data, and VIM gene methylation data. Circulating tumor DNA data includes ctDNA mutation detection data and ctDNA methylation detection data. Microbiome data includes gut microbiota analysis data.

[0026] Understandably, the health sample testing data includes data on various colorectal cancer-related biomarkers. Together, these data provide comprehensive and accurate support for the early diagnosis and risk assessment of colorectal cancer, helping to improve the accuracy and reliability of diagnosis.

[0027] In one possible implementation, after S1 and before S2, it further includes:

[0028] Standardize the testing data of healthy samples.

[0029] Specifically, standardization transforms the health sample test data so that the mean of each data feature is 0 and the standard deviation is 1. The specific method is as follows: for each data point, subtract the mean of that feature and then divide by the standard deviation of that feature. This transforms all data to the same scale, avoiding the influence of dimensional differences between different features on the model results and ensuring that all data are compared on the same scale.

[0030] S2: Based on the detection data of each group of healthy samples, determine the steady-state center representing the health status, and establish the Jacobian matrix describing the health status through a regression fitting process.

[0031] In this context, the steady-state center refers to a central point in the health sample data, identified by analyzing the average or representative state of all data, serving as a standard or benchmark for health status. Regression fitting is a mathematical method that uses the health sample data to find a pattern describing the relationships between health samples. The Jacobian matrix describes the sensitivity of health status to changes, representing the rate of change and degree of influence of health status relative to different variables. Here, the Jacobian matrix is ​​used to capture the non-linear relationship between health status and data. By combining regression fitting with the Jacobian matrix, the changing patterns of the health sample data can be accurately quantified, the steady-state center can be found, providing a scientific and stable benchmark for subsequent risk assessment, avoiding subjective biases, and ensuring higher accuracy.

[0032] In one possible implementation, S2 specifically includes:

[0033] S201: Calculate the mean value of each colorectal cancer marker in the test data of each group of healthy samples, and form a mean vector from the mean values ​​of various colorectal cancer markers. The mean vector is the center of steady state.

[0034] S202: Establish a training set based on the detection data and mean vector of each group of healthy samples. The training set includes the state deviation and the rate of change of the state deviation between the detection data and mean vector of each group of healthy samples.

[0035] The specific formula for the training set is as follows:

[0036] .

[0037] .

[0038] in, This represents the vector consisting of the detection data of the k-th group of healthy samples at time t. With mean vector State deviation between M represents the total number of healthy sample test data sets. This represents the rate of change of state deviation in the test data of the k-th group of healthy samples. This indicates the sampling interval for the same group of healthy sample test data.

[0039] It's important to note that the training set is constructed by calculating the deviation between the health sample detection data and the mean vector (the center of steady state). The difference between each set of health sample data and the mean vector is called the state deviation. Furthermore, the rate of change of these deviations over time is calculated, reflecting how the data changes over time. This training set allows for a better capture of the dynamic characteristics of health data, providing the necessary foundation for model building. The combination of these deviations and rates of change provides crucial information for subsequent analysis and prediction, helping to assess the relationship between health status and potential risks.

[0040] S203: Combining state deviation and the rate of change of state deviation, establish a linear regression equation including the Jacobian matrix based on the training set.

[0041] The linear regression equation is as follows:

[0042] .

[0043] in, This represents the Jacobian matrix.

[0044] S204: Establish a loss function with the objective of minimizing the rate of change of predicted state deviation and the rate of change of actual state deviation.

[0045] The loss function is as follows:

[0046] .

[0047] in, Let A represent the loss function with respect to the Jacobian matrix A. This represents the square of the L2 norm.

[0048] S205: Take the derivative of the loss function and set the derivative to zero to obtain the Jacobian matrix.

[0049] The Jacobian matrix is ​​specifically as follows:

[0050] .

[0051] In this context, the subscript T indicates transpose.

[0052] Specifically, this step involves establishing the deviation and rate of change of the deviation between healthy sample data and the steady-state center (mean vector), and then using linear regression to solve for the Jacobian matrix. First, a training set is constructed by calculating the state deviation and rate of change between each healthy sample data point and the steady-state center. Next, the Jacobian matrix is ​​constructed using linear regression on the training set data; this matrix describes the changing patterns of health status. The Jacobian matrix is ​​optimized by minimizing the error between the predicted and actual rates of change. Finally, the formula for calculating the Jacobian matrix is ​​based on the least squares method; the optimal Jacobian matrix is ​​obtained by minimizing the loss function. The Jacobian matrix accurately describes the dynamic changes in health status and captures the trends in healthy sample data, thus providing a reliable mathematical model for subsequent cancer risk assessment. Using linear regression and the least squares method, the Jacobian matrix can be obtained efficiently and accurately, avoiding complex nonlinear analysis while ensuring the interpretability and stability of the model.

[0053] In one possible implementation, after S205, the following is also included:

[0054] S206: Perform eigenvalue real part verification on the Jacobian matrix. If the verification passes, output the Jacobian matrix; otherwise, reacquire the health sample detection data and return to step S201.

[0055] The verification formula is as follows:

[0056] .

[0057] in, Let i represent the i-th eigenvalue of the Jacobian matrix A. express The real part.

[0058] It's important to note that the stability of the Jacobian matrix is ​​ensured by verifying the real parts of its eigenvalues. The verification formula requires that the real parts of all eigenvalues ​​of the Jacobian matrix are less than zero, meaning that the system's stability is guaranteed and there are no unstable or divergent solutions. If the verification passes, the obtained Jacobian matrix is ​​valid and can be used for subsequent analysis. Otherwise, health sample data is reacquired to ensure data accuracy and model stability, thus accurately describing the health status. The advantage of this step is that mathematical verification ensures the solved Jacobian matrix meets the stability requirements, thereby improving the reliability of the entire model and avoiding erroneous judgments caused by unstable matrices.

[0059] S3: Construct the Lyapunov equations using the Jacobian matrix as input, and solve for the positive definite matrix of the Lyapunov equations.

[0060] The Lyapunov equations are mathematical equations commonly used to analyze the stability of dynamic systems. By describing the relationship between system state changes and stability, they help determine whether a system will deviate from its steady state. They can predict system behavior by solving specific matrix equations. A positive definite matrix is ​​a matrix whose eigenvalues ​​are all positive, possessing favorable mathematical properties and often used to represent system stability and energy conservation. Positive definite matrices ensure system stability and avoid phenomena such as energy dissipation or infinity. By constructing a stability analysis model using the Lyapunov equations and solving for positive definite matrices, the stability relationship between healthy and abnormal states can be effectively captured. This method helps assess the stability of data, ensuring that fluctuations and risks in health states can be accurately quantified during detection, thereby improving the accuracy of colorectal cancer risk assessment.

[0061] In one possible implementation, the Lyapunov equations are specifically:

[0062] .

[0063] in, This represents the Jacobian matrix, with the subscript T indicating transpose. Describes a positive definite matrix. Represents the identity matrix.

[0064] It should be noted that the Lyapunov equations describe the stability of a system by introducing a positive definite matrix and an identity matrix. Solving the equations yields a positive definite matrix that characterizes the system's stability and energy dissipation properties. The advantage of this equation lies in its ability to effectively capture the system's stability, ensuring that appropriate matrix analysis methods are used to determine whether the system will diverge, avoiding the occurrence of unstable solutions, and thus enhancing the reliability and accuracy of system analysis.

[0065] S4: Establish a Lyapunov energy function based on a positive definite matrix to quantify the detection data as the energy cost of deviating from the steady-state center.

[0066] The Lyapunov energy function, based on the Lyapunov equations, describes the energy cost of a system's state deviating from its steady-state center. It quantifies the system's dynamic changes as energy, representing the degree of deviation from the steady-state center. The energy function helps measure the system's deviation from its ideal healthy state, thus predicting system stability. By quantifying the deviation of data using the Lyapunov energy function, the difference between healthy and abnormal states can be clearly presented in the form of energy. This quantification method provides an objective and quantitative standard, avoiding the subjectivity of traditional manually set thresholds, thereby improving the accuracy and reliability of colorectal cancer risk assessment.

[0067] S5: Obtain the test data for the colorectal cancer kit.

[0068] The data to be tested comes from biomarkers detected by the colorectal cancer kit and is presented as a vector. Each element represents the detection value of a specific biomarker, reflecting biological changes in the patient's body. Analyzing this biomarker data can help determine whether a patient is at risk of colorectal cancer.

[0069] S6: Input the data to be detected into the Lyapunov energy function and solve for the rate of change of the Lyapunov energy function.

[0070] The Lyapunov energy function rate of change refers to the rate at which the energy function value changes with time or a variable after the data to be tested is input into the Lyapunov energy function. By calculating the rate of change of the Lyapunov energy function, the degree of deviation between the data to be tested and the state of health can be accurately measured.

[0071] In one possible implementation, the Lyapunov energy function is specifically:

[0072] .

[0073] in, Indicates the data to be detected The Lyapunov energy function value, This represents the mean vector, i.e., the steady-state center, with the subscript T indicating transpose. This represents a positive definite matrix.

[0074] It should be noted that the Lyapunov energy function represents the degree of deviation between the data to be tested and the steady-state center. By calculating the difference between the data to be tested and the steady-state center, and combining this with a positive definite matrix, the energy function quantifies the energy cost of the deviation. The positive definite matrix is ​​used to ensure system stability and energy dissipation characteristics. The advantage of this energy function is that it can accurately quantify the degree of deviation of the data, which helps to determine whether the data is consistent with a healthy state and improves the accuracy of colorectal cancer risk assessment.

[0075] In one possible implementation, the solution for the rate of change of the Lyapunov energy function in S6 is specifically as follows:

[0076] By combining the Lyapunov equations and deriving the rate of change of the Lyapunov energy function based on stability theory, we can obtain the result.

[0077] The derivation formula for the rate of change of Lyapunov's energy function is as follows:

[0078] .

[0079] in, Indicates the data to be detected The rate of change of Lyapunov's energy function at time t.

[0080] It's important to note that the derivation of the Lyapunov energy function's rate of change utilizes stability theory. It determines system stability by calculating the difference between the data being tested and the steady-state center. The rate of change in the formula represents the system's energy change. When the data deviates from the healthy steady-state center, the energy increases, resulting in a positive rate of change, indicating potential instability or anomalies in the system. Conversely, a negative rate of change indicates a decrease in system energy, suggesting the data is approaching a healthy state and indicating no risk. This process allows for a more accurate assessment of colorectal cancer risk, ensuring the stability and reliability of the test results.

[0081] S7: If the rate of change of the Lyapunov energy function is less than zero, the output indicates no risk of colorectal cancer; otherwise, the output indicates a risk of colorectal cancer.

[0082] Understandably, the risk of colorectal cancer is assessed by analyzing the rate of change of the Lyapunov energy function. If the rate of change is less than zero, it means the deviation between the tested data and a healthy state is small, indicating system stability and no risk of colorectal cancer. If the rate of change is greater than zero, it indicates that the data deviates significantly from a healthy state, potentially suggesting abnormal changes or colorectal cancer risk. This method allows for an objective and accurate assessment of cancer risk, reducing subjective interference and improving diagnostic precision.

[0083] In one possible implementation, after S7, the following is also included:

[0084] The Lyapunov energy function is updated at preset intervals.

[0085] It should be noted that those skilled in the art can set the preset duration according to actual needs, and this invention does not impose any limitations on it. Updating the Lyapunov energy function at preset intervals is to continuously track the deviation between the data to be detected and the health status, ensuring that the model can reflect data changes in real time and provide updated risk assessments. This helps improve the accuracy of diagnosis in long-term monitoring.

[0086] In practical applications, this scheme achieves colorectal cancer detection by integrating machine learning and stability theory. Specifically, firstly, using multi-dimensional biomarker data from healthy individuals, a steady-state center is determined through standardization and mean calculation. Then, a Jacobian matrix is ​​fitted using linear regression to capture the negative feedback regulation of the healthy system, ensuring that the real parts of the matrix eigenvalues ​​are negative to guarantee stability. Next, the Lyapunov equation is solved to obtain a positive definite matrix, and an energy function is constructed to quantify the "energy cost" of the data deviating from steady state. Finally, the risk is judged by the sign of the energy change rate (without the need for a manual threshold): a change rate < 0 indicates that the system has converged to a healthy steady state (no risk), while a change rate > 0 indicates divergence (risk). This approach breaks through the limitations of traditional static thresholds, transforming physiological states into stability problems and capturing the essence of homeostasis disruption caused by cancer. The Jacobian matrix is ​​fitted using healthy data (machine learning), and the energy function is based on rigorous stability theory (Lyapunov method), combining data adaptability with mathematical rigor. The positive definite matrix automatically quantizes feature weights, and uses the sign of the energy change rate as the criterion to avoid subjective threshold errors. In addition, it supports input of multiple omics markers, and the matrix coupling characteristics naturally integrate feature correlations, improving the robustness and reliability of detection for complex data.

[0087] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following:

[0088] In this embodiment of the invention, a steady-state center of health is determined by regression fitting of healthy sample data, and a Jacobian matrix is ​​established to describe the changing characteristics of health status. Then, a Lyapunov equation is constructed based on the Jacobian matrix and solved to obtain a positive definite matrix, thereby establishing a Lyapunov energy function to quantify the degree to which the detection data deviates from the healthy steady-state center. After the data to be detected is input into the Lyapunov energy function, the calculated rate of change reflects the degree of deviation of the data relative to the healthy steady-state center. If the rate of change is less than zero, it indicates that the deviation between the data to be detected and the healthy state is small, the data is in a stable state, and no abnormalities have occurred, therefore there is no risk of colorectal cancer. Conversely, if the rate of change is greater than zero, it indicates that the data to be detected deviates from the healthy steady-state center, and there is a large change or abnormality, suggesting a possible risk of colorectal cancer. This process, by accurately quantifying data deviation, avoids subjective interference from manually set thresholds, making risk assessment more objective and accurate. It improves the objectivity and accuracy of the detection results, avoids the setting of manually set thresholds, and can more accurately identify potential risks.

[0089] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the embodiments of the present invention, and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the protection scope of the present invention.

Claims

1. A machine learning-driven intelligent processing method for colorectal cancer reagent kit detection data, characterized in that, include: S1: Obtain multiple sets of test data for healthy samples obtained from the colorectal cancer reagent kit; S2: Based on the detection data of the health samples in each group, determine the steady-state center representing the health status, and establish the Jacobian matrix describing the health status through a regression fitting process; S3: Construct the Lyapunov equation using the Jacobian matrix as input, and solve for the positive definite matrix of the Lyapunov equation; S4: Establish a Lyapunov energy function based on the positive definite matrix to quantify the detection data as the energy cost of deviating from the steady-state center; S5: Obtain the test data of the colorectal cancer reagent kit; S6: Input the data to be detected into the Lyapunov energy function and solve for the rate of change of the Lyapunov energy function; S7: If the rate of change of the Lyapunov energy function is less than zero, the output indicates no risk of colorectal cancer; otherwise, the output indicates a risk of colorectal cancer.

2. The machine learning-driven intelligent data processing method for colorectal cancer reagent kit detection according to claim 1, characterized in that, The health sample test data includes multiple colorectal cancer biomarker data, including tumor biomarker data, gene mutation data, DNA methylation data, circulating tumor DNA data, and microbiome data.

3. The machine learning-driven intelligent data processing method for colorectal cancer reagent kit detection according to claim 2, characterized in that, The tumor marker data includes carcinoembryonic antigen (CEA) data, CA19-9 data, M2-PK data, and CA72-4 data; the gene mutation data includes KRAS gene mutation data, BRAF gene mutation data, TP53 gene mutation data, and PIK3CA gene mutation data; the DNA methylation data includes MLH1 gene methylation data, CDKN2A gene methylation data, and VIM gene methylation data; the circulating tumor DNA data includes ctDNA mutation detection data and ctDNA methylation detection data; and the microbiome data includes gut microbiota analysis data.

4. The machine learning-driven intelligent data processing method for colorectal cancer reagent kit detection according to claim 1, characterized in that, After S1 and before S2, it also includes: The health sample test data are standardized.

5. The machine learning-driven intelligent data processing method for colorectal cancer reagent kit detection according to claim 1, characterized in that, S2 specifically includes: S201: Calculate the mean value of each colorectal cancer marker data in the health sample detection data of each group, and form a mean vector from the mean values ​​of various colorectal cancer marker data, wherein the mean vector is the steady state center; S202: Establish a training set based on the health sample detection data of each group and the mean vector, wherein the training set includes the state deviation and the rate of change of the state deviation between the health sample detection data of each group and the mean vector; S203: Combining the state deviation and the rate of change of the state deviation, establish a linear regression equation including the Jacobian matrix based on the training set; S204: Establish a loss function with the objective of minimizing the rate of change of predicted state deviation and the rate of change of actual state deviation; S205: Take the derivative of the loss function and set the derivative to zero to obtain the Jacobian matrix.

6. The machine learning-driven intelligent data processing method for colorectal cancer reagent kit detection according to claim 5, characterized in that, Following S205, the following is also included: S206: Perform eigenvalue real part verification on the Jacobian matrix. If the verification passes, output the Jacobian matrix; otherwise, reacquire the health sample detection data and return to step S201.

7. The machine learning-driven intelligent data processing method for colorectal cancer reagent kit detection according to claim 1, characterized in that, The Lyapunov equations are specifically as follows: ； in, This represents the Jacobian matrix, with the subscript T indicating transpose. Describes a positive definite matrix. Represents the identity matrix.

8. The machine learning-driven intelligent processing method for colorectal cancer reagent kit detection data according to claim 1, characterized in that, The Lyapunov energy function is specifically as follows: ； in, Indicates the data to be detected The Lyapunov energy function value, This represents the mean vector, i.e., the steady-state center, with the subscript T indicating transpose. This represents a positive definite matrix.

9. The machine learning-driven intelligent processing method for colorectal cancer reagent kit detection data according to claim 1, characterized in that, Specifically, the solution for the rate of change of the Lyapunov energy function in S6 is as follows: Based on the Lyapunov equation and stability theory, the rate of change of the Lyapunov energy function is derived.

10. A machine learning-driven intelligent processing method for colorectal cancer reagent kit detection data, characterized in that, Following S7, it also includes: The Lyapunov energy function is updated at preset intervals.

Images