An intelligent system for evaluating the effect of immune cell therapy

Abstract

The present invention relates to the field of medical information processing, and in particular to an intelligent system for evaluating the effect of immune cell therapy. The content includes: collecting and preprocessing raw immune index data to obtain immune index data; driving the evaluation algorithm through adaptive fluctuation entropy to obtain the adaptive fluctuation weight and dynamic regular information entropy of the immune index data, and calculating the cumulative state trend value; based on the immune index data and the cumulative state trend value, calculating the total error and the error change, evaluating the immediate treatment effect and generating a monitoring log; dividing the time window, calculating the stability index of the time window, and evaluating the treatment effect. It solves the problem that most traditional treatment effect evaluation methods cannot sensitively capture the subtle fluctuations of immune indicators and it is difficult to timely warn potential adverse reactions; most of them are based on static or single weight allocation methods, which are easy to miss the real-time fluctuations of the treatment effect; most of them cannot track the cumulative state trend of data and it is difficult to provide continuous efficacy evaluation.

Description

Technical Field

[0001] The present invention relates to the field of medical information processing, and in particular to an intelligent system for evaluating the therapeutic effect of immune cells. Background Art

[0002] In recent years, with the deepening of the application of immune cell therapy in the fields of tumors, infectious diseases and autoimmune diseases, the objective evaluation of treatment effects has gradually become an important focus of clinical and research. In the existing technology, the evaluation of treatment effects usually relies on the analysis of multidimensional data, including changes in the concentration of biomarkers in the patient's blood, imaging information and real-time clinical observation records. In order to have a more comprehensive understanding of the immune response, researchers have gradually adopted data-driven models, using biomarkers, cell activity and other indicators as evaluation criteria, and combined with imaging changes to evaluate the efficacy.

[0003] However, due to the complexity and individual differences of immunotherapy, how to integrate and analyze multi-source data, capture changes in patients' immune responses in real time, and generate dynamic and objective efficacy evaluation reports is still a difficult point in clinical evaluation. First, traditional treatment effect evaluation methods generally lack the ability to sensitively capture subtle fluctuations in immune indicators, which makes it difficult to timely warn of potential adverse reactions during complex treatment processes; second, traditional treatment effect evaluation methods are mostly based on static or single weight distribution methods, which easily miss real-time fluctuations in treatment effects; in addition, traditional treatment effect evaluation methods mostly lack the ability to track trends in data accumulation status, making it difficult to provide continuous efficacy evaluations, making it difficult for clinicians to effectively monitor the stability and durability of treatment. The above problems result in the inability of traditional treatment effect evaluation methods to provide sufficient data support and early warning information for the optimization of individualized treatment plans, and they urgently need to be improved through a new algorithm and processing technology. Summary of the invention

[0004] The present invention provides an intelligent system for evaluating the therapeutic effects of immune cells, so as to solve the problems that traditional therapeutic effect evaluation methods generally lack the sensitivity to capture subtle fluctuations in immune indicators, resulting in difficulty in timely warning of potential adverse reactions during complex treatment processes; most of them are based on static or single weight allocation methods, which are prone to missing real-time fluctuations in therapeutic effects; most of them lack the ability to track the trend of data accumulation status, making it difficult to provide continuous efficacy evaluation, making it difficult for clinicians to effectively monitor the stability and durability of treatment.

[0005] The intelligent system for evaluating the effect of immune cell therapy of the present invention specifically includes the following technical solutions:

[0006] An intelligent method for evaluating the effect of immune cell therapy comprises the following steps:

[0007] S1: Collect the original immune indicator data and perform preprocessing to obtain the immune indicator data; based on the immune indicator data, the adaptive fluctuation weight and dynamic regular information entropy of the immune indicator data are obtained by driving the evaluation algorithm through the adaptive fluctuation entropy; based on the dynamic regular information entropy, the cumulative state trend value is calculated;

[0008] S2: Based on the immune indicator data and the cumulative state trend value, calculate the total error and the error change; based on the error change, evaluate the immediate treatment effect and generate a monitoring log; divide the time window, and calculate the stability index of the time window based on the cumulative state trend value, the total error and the error change; based on the stability index of the time window, evaluate the treatment effect and generate a comprehensive treatment evaluation report.

[0009] Preferably, the S1 specifically includes:

[0010] The adaptive fluctuation entropy driven evaluation algorithm introduces adaptive fluctuation weights and dynamic regular information entropy to perform real-time analysis on immune indicator data.

[0011] Preferably, the S1 specifically includes:

[0012] In the implementation process of the adaptive fluctuation entropy driven evaluation algorithm, based on the immune indicator data, sine and cosine are introduced for nonlinear adjustment, and the adaptive fluctuation weight is calculated.

[0013] Preferably, the S1 specifically includes:

[0014] Based on the adaptive fluctuation weight, the dynamic adjustment term and regularization parameter are introduced to calculate the dynamic regularized information entropy. The specific formula is as follows:

[0015]

[0016] Where H(t) is the dynamic regular information entropy at time t; i is the index variable of the immune indicator data; n is the total number of immune indicator data; p i (t) is the adaptive fluctuation weight of the i-th immune indicator data at time t; p i (t-1) is the adaptive fluctuation weight of the ith immune indicator data at time t-1; ∈ is a constant used to avoid zero values ​​in logarithmic calculation; α is the dynamic change adjustment coefficient; represents the dynamic adjustment term; η is the regularization parameter.

[0017] Preferably, the S1 specifically includes:

[0018] The dynamic regular information entropy at each time is recursively updated to construct the state trend that changes over time and obtain the cumulative state trend value.

[0019] Preferably, the S2 specifically includes:

[0020] Construct a time series model, analyze the patient's historical immune indicator data, and obtain the time series model prediction value; calculate the deviation between the immune indicator data and the time series model prediction value, and combine the cumulative state trend value to obtain the total error; based on the total error, obtain the error change between adjacent time points.

[0021] Preferably, the S2 specifically includes:

[0022] Based on the divided time windows, the stability index of each time window is calculated by combining the total error, error change and cumulative state trend value within the time window. The specific formula is as follows:

[0023]

[0024] Among them, S k is in the kth time window [t k , t k+1 ] is the stability index; Z(t-1) is the cumulative state trend value at time t-1; ΔE(t) is the error change at time t; t k is the starting point of the kth time window; t k+1 is the end point of the kth time window; k is the index variable of the time window; E(t) is the total error at time t; Z(t) is the cumulative state trend value at time t; Z(t-1) is the cumulative state trend value at time t-1.

[0025] Preferably, the S2 specifically includes:

[0026] By performing trend analysis on the stability index of the time window, the time window of abnormal state is identified, and the stability index analysis results of all time windows are summarized to generate a comprehensive treatment evaluation report.

[0027] An intelligent system for evaluating the effect of immune cell therapy, comprising the following parts:

[0028] Data collection and preprocessing module, adaptive fluctuation weight module, dynamic regular information entropy module, cumulative state trend module, total error and error change module, stability index module;

[0029] The data collection and preprocessing module collects and preprocesses the original immune index data to obtain and output the immune index data; the data collection and preprocessing module is connected to the adaptive fluctuation weight module, the dynamic regular information entropy module, and the total error and error change module by means of data transmission;

[0030] The adaptive fluctuation weight module receives the immune index data, calculates the adaptive fluctuation weight and outputs it; the adaptive fluctuation weight module is connected to the dynamic regular information entropy module by means of data transmission;

[0031] The dynamic regular information entropy module receives the adaptive fluctuation weight and the immune index data, calculates the dynamic regular information entropy and outputs it; the dynamic regular information entropy module is connected to the cumulative state trend module by means of data transmission;

[0032] The cumulative state trend module receives the dynamic regular information entropy, recursively updates the dynamic regular information entropy at each time, constructs a state trend that changes over time, obtains the cumulative state trend value and outputs it; the cumulative state trend module is connected to the total error and error change module and the stability index module by means of data transmission;

[0033] The total error and error change module receives the cumulative state trend value and the immune index data, calculates the total error and the error change; based on the error change, evaluates the immediate treatment effect and generates a monitoring log; outputs the total error and the error change; the total error and error change module is connected to the stability index module by means of data transmission;

[0034] The stability index module receives the cumulative state trend value, the total error, and the error change, divides the treatment process into time windows, and calculates the stability index of the time window based on the total error, the error change, and the cumulative state trend value in the time window; by performing trend analysis on the stability index of the time window, the time window of abnormal state is identified, the stability index analysis results of all time windows are summarized, and a comprehensive treatment evaluation report is generated.

[0035] The beneficial effects of the technical solution of the present invention are:

[0036] 1. The present invention uses an adaptive fluctuation entropy driven evaluation algorithm, introduces adaptive fluctuation weights and dynamic regular information entropy, and captures the subtle fluctuations and nonlinear changes of each immune indicator data during immunotherapy with high precision; the adaptive fluctuation weights adjust the weights of each immune indicator data in real time, and combined with the sensitivity of dynamic regular information entropy, quickly respond to abnormal changes in immune indicator data. Compared with the traditional single weight allocation method, it improves the fluctuation recognition ability of immune indicator data, helps clinical physicians to identify treatment risks in a timely and accurate manner, and provides a reliable basis for personalized treatment adjustments.

[0037] 2. The present invention calculates the total error and the error change, evaluates the deviation between the patient's treatment status and the time series model prediction value in real time, provides instant feedback, and sensitively captures the deviation of the treatment effect through dynamic adjustment of the total error combined with the cumulative state trend value; the error change reflects the fluctuation amplitude of adjacent time points, helps to quickly identify anomalies, and when the error change is greater than the critical threshold, the automatic warning prompts the clinical physician to adjust the treatment plan in time to better meet the patient's needs and improve the safety and stability of treatment.

[0038] 3. The present invention divides multiple time windows and calculates the stability index of the time window to evaluate the stability and long-term trend of the treatment effect. The stability index of the time window comprehensively considers the total error, error change and cumulative state trend value within the time window to more comprehensively quantify the risk of treatment fluctuations. A high stability index indicates potential instability and provides physicians with accurate long-term risk warnings. Compared with traditional single time point evaluation, the present invention realizes continuous monitoring of the treatment process and provides data support for personalized treatment optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 This is a structural diagram of an intelligent system for evaluating the effect of immune cell therapy according to the present invention;

[0040] Figure 2 This is a flow chart of an intelligent method for evaluating the effect of immune cell therapy described in the present invention. DETAILED DESCRIPTION

[0041] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined purpose of the invention, the technical scheme in the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0042] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0043] The specific scheme of an intelligent system for evaluating the effect of immune cell therapy provided by the present invention is described in detail below with reference to the accompanying drawings.

[0044] See attached Figure 1 , which shows a structure diagram of an intelligent system for evaluating the effect of immune cell therapy provided by an embodiment of the present invention, the system includes the following parts:

[0045] Data collection and preprocessing module, adaptive fluctuation weight module, dynamic regular information entropy module, cumulative state trend module, total error and error change module, stability index module;

[0046] The data collection and preprocessing module collects the original immune index data of the patient (such as biomarker concentration, dynamic changes of immune cells, imaging detection data) and performs preprocessing to obtain and output the immune index data; the preprocessing process includes data cleaning, missing value processing, noise filtering, format conversion and normalization processing; the data collection and preprocessing module is connected to the adaptive fluctuation weight module, the dynamic regular information entropy module, and the total error and error change module by means of data transmission;

[0047] The adaptive fluctuation weight module receives the immune index data, calculates the adaptive fluctuation weight, realizes the dynamic adjustment of the importance of different immune index data in the overall evaluation, and outputs the adaptive fluctuation weight; the adaptive fluctuation weight module is connected to the dynamic regular information entropy module by means of data transmission;

[0048] The dynamic regular information entropy module receives the adaptive fluctuation weight and immune index data, calculates the dynamic regular information entropy, captures the state changes during the immunotherapy process, and outputs the dynamic regular information entropy; the dynamic regular information entropy module is connected to the cumulative state trend module by means of data transmission;

[0049] The cumulative state trend module receives the dynamic regular information entropy, recursively updates the dynamic regular information entropy at each time, constructs a state trend that changes over time, obtains a cumulative state trend value, provides a basis for efficacy evaluation, and outputs a cumulative state trend value; the cumulative state trend module is connected to the total error and error change module and the stability index module by means of data transmission;

[0050] The total error and error change module receives the cumulative state trend value and the immune index data, calculates the total error and the error change; based on the error change, evaluates the immediate treatment effect and generates a monitoring log, and outputs the total error and the error change; the total error and error change module is connected to the stability index module by means of data transmission;

[0051] The stability index module receives the cumulative state trend value, the total error, and the error change, divides the treatment process into multiple time windows, and calculates the stability index of the time window based on the total error, the error change, and the cumulative state trend value in the time window; by performing trend analysis on the stability index of the time window, the time window of abnormal state is identified, the stability index analysis results of all time windows are summarized, and a comprehensive treatment evaluation report is generated.

[0052] See attached Figure 2, which shows a flow chart of an intelligent method for evaluating the effect of immune cell therapy provided by an embodiment of the present invention, the method comprising the following steps:

[0053] S1: Collect the original immune indicator data and perform preprocessing to obtain the immune indicator data; based on the immune indicator data, the adaptive fluctuation weight and dynamic regular information entropy of the immune indicator data are obtained by driving the evaluation algorithm through the adaptive fluctuation entropy; based on the dynamic regular information entropy, the cumulative state trend value is calculated;

[0054] Collect the original immune index data of patients, including biomarker concentrations in the patient's blood (such as cytokine levels, white blood cell and lymphocyte counts), dynamic changes in immune cells (such as changes in the number of specific cells over time), imaging detection data (such as lesion size and distribution), and clinical observation records during treatment (such as adverse reactions, temperature changes, etc.). The original immune index data comes from a variety of equipment and detection methods, such as laboratory test reports, medical imaging systems, and patient health monitoring equipment. Set the type of original immune index data according to the specific implementation scenario, and collect it according to a unified time node to ensure the matching of original immune index data in different time periods.

[0055] The original immune index data is pre-processed to obtain immune index data; pre-processing includes data cleaning, missing value processing, noise filtering, format conversion and normalization. Specifically, repeated or erroneous original immune index data are eliminated by data cleaning to obtain the original immune index data after data cleaning; the original immune index data after data cleaning is supplemented with missing values ​​by interpolation, mean filling or regression prediction to obtain the original immune index data after missing value processing; by smoothing or nonlinear filtering, the original immune index data after missing value processing is subjected to noise filtering to ensure data stability, and the original immune index data after noise filtering is obtained; the time format and unit dimension of the original immune index data after noise filtering are unified by format conversion to obtain the original immune index data after format conversion; the original immune index data after format conversion is scaled to the same numerical range by normalization to obtain immune index data. The data collection and pre-processing techniques are well known to those skilled in the art, and will not be repeated here.

[0056] The adaptive fluctuation entropy driven evaluation algorithm is used to analyze the immune indicator data in real time to capture the subtle fluctuations and nonlinear changes of each immune indicator data during immune cell therapy. Specifically, the adaptive fluctuation entropy driven evaluation algorithm introduces adaptive fluctuation weights and dynamic regular information entropy to capture the state fluctuations of immune indicator data during treatment, and generates a cumulative state trend value that reflects the overall trend of immune indicator data for continuous monitoring of immune response.

[0057] The adaptive fluctuation weight is calculated by the adaptive fluctuation entropy driven evaluation algorithm based on the fluctuation amplitude and nonlinear changes of immune indicator data at different time points; the dynamic regularized information entropy is calculated by introducing dynamic adjustment terms and regularization parameters on the basis of traditional information entropy, and is used to accurately capture state fluctuations and sudden changes during the treatment process; the cumulative state trend value is a recursive update of the dynamic regularized information entropy at each time to construct a state trend that changes over time, which is used to track the efficacy trend throughout the process and assist in long-term effect prediction.

[0058] The specific process of the adaptive fluctuation entropy driven evaluation algorithm is as follows:

[0059] First, the adaptive fluctuation weight is calculated to dynamically adjust the importance of immune index data in the overall evaluation and enhance the sensitivity to subtle fluctuations during treatment. The adaptive fluctuation weight formula is as follows:

[0060]

[0061] Among them, p i (t) is the adaptive fluctuation weight, which represents the normalized weight of the i-th immune index data at time t, and is used to describe the importance of the immune index data at the current time; i is the index variable of the immune index data, corresponding to the specific immune index data; t is the time variable, which represents the time point of data collection; x i (t) is the i-th immune indicator data at time t; β is a nonlinear adjustment parameter used to control the influence of sine and cosine terms on the adaptive fluctuation weight calculation, which is obtained through experimental tuning; n is the total number of immune indicator data, which is set according to the specific implementation scenario; j is the index variable of the immune indicator data in the summation operation; x j (t) is the,th immune index data at time t;

[0062] The adaptive fluctuation weight formula introduces sine and cosine for nonlinear adjustment, which improves the sensitivity of the adaptive fluctuation weight to subtle fluctuations in immune indicator data and increases the accuracy of dynamic state assessment.

[0063] Furthermore, the adaptive fluctuation weight is used in the calculation of dynamic regular information entropy, so that the dynamic regular information entropy can accurately capture the state changes of immune indicator data during immunotherapy. The dynamic regular information entropy formula is as follows:

[0064]

[0065] Where H(t) is the dynamic regularized information entropy at time t, which is used to measure the state complexity in the immune cell therapy process; p i(t-1) is the adaptive fluctuation weight of the ith immune indicator data at time t-1; ∈ is a constant used to avoid zero values ​​in logarithmic calculations, which is set according to the specific implementation scenario; α is a dynamic change adjustment coefficient, which is used to control the impact of the change in the adaptive fluctuation weight between the current time and the previous time on the dynamic regular information entropy, and is obtained through experimental tuning; represents the dynamic adjustment term; η is the regularization parameter, which is used to prevent abnormal values ​​of immune indicator data from excessively affecting the calculation results of dynamic regularized information entropy and is obtained through experimental tuning;

[0066] The dynamic regularized information entropy formula introduces dynamic adjustment terms and regularization parameters, integrates the changes in the adaptive fluctuation weights of the current time and the previous time and the abnormal value control of the immune index data into the dynamic regularized information entropy calculation to measure the complexity fluctuations in the immunotherapy process;

[0067] Finally, the cumulative state trend value is constructed based on the dynamic regular information entropy. The cumulative state trend value is adjusted according to the change trend of the dynamic regular information entropy to reflect the cumulative changes in the treatment effect during immunotherapy in real time, which serves as the basis for real-time adjustment of the treatment process. The calculation formula of the cumulative state trend value is as follows:

[0068]

[0069] Among them, Z(t) is the cumulative state trend value at time t, which represents the cumulative overall treatment state up to time point t. The cumulative state trend value is dynamically recursive over time and is used to describe the comprehensive state of the intelligent system for evaluating the effect of immune cell therapy during immunotherapy; Z(t-1) is the cumulative state trend value at time t-1, that is, the cumulative state trend value at the previous time; the initial value of the cumulative state trend value Z(0) = 0; H(t-1) is the dynamic regular information entropy at time t-1;

[0070] The cumulative state trend formula accumulates the state changes of immune indicator data before and after into continuous state trend values ​​through the recursive mechanism of dynamic regular information entropy, so that the state evolution during immunotherapy can be tracked and quantified in real time.

[0071] S2: Based on the immune indicator data and the cumulative state trend value, calculate the total error and the error change; based on the error change, evaluate the immediate treatment effect and generate a monitoring log; divide the time window, and calculate the stability index of the time window based on the cumulative state trend value, the total error and the error change; based on the stability index of the time window, evaluate the treatment effect and generate a comprehensive treatment evaluation report.

[0072] Based on the cumulative state trend value, the total error of the corresponding time is calculated to reflect the immediate gap between the current treatment state and the expected model, and promptly prompt clinicians whether they need to adjust the treatment plan according to the actual situation of individual patients.

[0073] The total error is used to quantify the deviation between the immune index data and the predicted value of the time series model, and is used to evaluate the adaptability and adjustment needs of the treatment plan in real time, and to assist in evaluating the efficacy; the time series model is established based on the patient's historical immune index data, and the patient's historical immune index data is analyzed based on the time series model to obtain the model predicted value, and the model predicted value is used as a comparison benchmark for the actual treatment effect to predict the ideal state of the immune index data during the treatment process; the time series model is a prior art and will not be described in detail here;

[0074] The total error is calculated as follows:

[0075]

[0076] Where E(t) is the total error at time t, which quantifies the deviation between the immune indicator data and the model prediction value. The higher the total error value, the greater the gap between the time series model and the actual performance; t is the time variable; Z(t) is the cumulative state trend value at time t; n is the total number of immune indicator data; i is the index variable of the immune indicator data; x i (t) is the i-th immune index data at time t; is the model predicted value of the ith immune index data at time t; is the immune index deviation of the ith immune index data at time t;

[0077] The total error calculation formula combines the immune index deviation with the cumulative state trend value, which enhances the sensitivity of the total error to the fluctuation of the cumulative state trend value and quantifies the gap between the model prediction value and the actual immune index data;

[0078] Furthermore, the error variation between adjacent time points is calculated to identify whether there are abnormal fluctuations during the treatment process. By analyzing the error variation, potential risks or emergencies during the treatment process can be quickly captured, providing more timely feedback to clinicians and ensuring the adaptability and safety of the treatment plan;

[0079] Specifically, the critical threshold is set using expert experience. When the error change is greater than the critical threshold, the current state is judged to be a critical state, and a warning signal is immediately triggered to remind the clinician that the patient may be at increased risk or the treatment effect may be deviated. The warning signal serves as the basis for adjusting the treatment plan, reminding the physician to make personalized adjustments to the dosage, frequency or medication method of the treatment plan to prevent further development of potential adverse reactions. After the warning signal is triggered, the time point of the critical state and the specific error value are recorded, and a monitoring log is generated and recorded in the patient's treatment file for subsequent analysis of the patient's treatment adaptability, providing data support for the optimization of long-term efficacy;

[0080] Furthermore, in order to measure the overall stability of the treatment process, the expert experience method is used to divide the treatment process into multiple windows [t k , t k+1 ], where t k is the starting point of the kth time window, t k+1 It is the end point of the kth time window and the starting point of the k+1th time window. k is the index variable of the time window.

[0081] Based on the divided time windows, the stability index of each time window is calculated; the stability index of the time window comprehensively reflects the stability of the treatment effect and the persistence of the immune response within the time window by considering the total error and the error change within the time window, combined with the cumulative state trend value, and can be used for subsequent analysis of the persistence of the treatment effect and possible fluctuation risks, providing the clinic with a long-term basis and early warning information for adjusting the treatment plan;

[0082] The specific formula of the stability index is as follows:

[0083]

[0084] Among them, S k is in the kth time window [t k , t k+1 ] is the stability index; Z(t-1) is the cumulative state trend value at time t-1; ΔE(t) is the error change at time t;

[0085] The stability index formula quantifies the stability of the treatment process in different time windows by comprehensively considering the total error, error change, and cumulative state trend value, providing a more sensitive immune indicator for evaluating immune response. Specifically, a high stability index indicates that the treatment effect fluctuates greatly within the time window, indicating that the treatment is unstable;

[0086] Furthermore, after calculating the stability index of each time window, the stability index is compared with a preset stability threshold. When the stability index of a time window is greater than the stability threshold, the time window is marked as an abnormal state, prompting the clinician that there is a treatment fluctuation or risk. Then, a trend analysis is performed on the stability index of all windows to determine the direction and magnitude of the state change. If the stability index continues to rise, it indicates that the treatment stability may decrease, prompting the clinician to increase the monitoring frequency or adjust the treatment plan.

[0087] The stability index analysis results of all time windows are summarized to generate a comprehensive treatment evaluation report, which includes the stability index, total error, error change, cumulative state trend value and time window for marking abnormal states for each window. The comprehensive treatment evaluation report is recorded in the patient's treatment file as a reference for clinical physicians to adjust treatment plans and risk warnings, providing data support for subsequent personalized treatment plans.

[0088] In summary, an intelligent system for evaluating the effects of immune cell therapy has been completed.

[0089] The order of the embodiments of the invention is for description only and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0090] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[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 above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.

Claims

1. An intelligent method for evaluating the effect of immune cell therapy, characterized in that: The following steps are involved: S1: Collect and preprocess the original immune index data to obtain immune index data; Based on the immune indicator data, the adaptive fluctuation weight and dynamic regular information entropy of the immune indicator data are obtained through the adaptive fluctuation entropy driven evaluation algorithm; Based on the dynamic regular information entropy, the cumulative state trend value is calculated; S2: Calculate the total error and error change based on the immune index data and cumulative state trend value; Based on the error change, evaluate the immediate treatment effect and generate a monitoring log; Divide the time window and calculate the stability index of the time window based on the cumulative state trend value, total error and error change; Based on the stability index of the time window, the treatment effect is evaluated and a comprehensive treatment evaluation report is generated.

2. An intelligent method for evaluating the effect of immune cell therapy according to claim 1, characterized in that: The S1 specifically includes: The adaptive fluctuation entropy driven evaluation algorithm introduces adaptive fluctuation weights and dynamic regular information entropy to perform real-time analysis on immune indicator data.

3. An intelligent method for evaluating the effect of immune cell therapy according to claim 2, characterized in that: The S1 specifically includes: In the implementation process of the adaptive fluctuation entropy driven evaluation algorithm, based on the immune indicator data, sine and cosine are introduced for nonlinear adjustment, and the adaptive fluctuation weight is calculated.

4. The intelligent method for evaluating the effect of immune cell therapy according to claim 3, characterized in that: The S1 specifically includes: Based on the adaptive fluctuation weight, the dynamic adjustment term and regularization parameter are introduced to calculate the dynamic regularized information entropy. The specific formula is as follows: Where H(t) is the dynamic regular information entropy at time t; i is the index variable of the immune indicator data; n is the total number of immune indicator data; p i (t) is the adaptive fluctuation weight of the i-th immune indicator data at time t; p i (t-1) is the adaptive fluctuation weight of the ith immune indicator data at time t-1; ∈ is a constant used to avoid zero values ​​in logarithmic calculation; α is the dynamic change adjustment coefficient; represents the dynamic adjustment term; η is the regularization parameter.

5. The intelligent method for evaluating the effect of immune cell therapy according to claim 4, characterized in that: The S1 specifically includes: The dynamic regular information entropy at each time is recursively updated to construct the state trend that changes over time and obtain the cumulative state trend value.

6. The intelligent method for evaluating the effect of immune cell therapy according to claim 1, characterized in that: The S2 specifically includes: Construct a time series model, analyze the patient's historical immune indicator data, and obtain the time series model prediction value; calculate the deviation between the immune indicator data and the time series model prediction value, and combine the cumulative state trend value to obtain the total error; based on the total error, obtain the error change between adjacent time points.

7. An intelligent method for evaluating the effect of immune cell therapy according to claim 6, characterized in that: The S2 specifically includes: Based on the divided time windows, the stability index of each time window is calculated by combining the total error, error change and cumulative state trend value within the time window. The specific formula is as follows: Among them, S k is in the kth time window [t k , t k+1 ] is the stability index; Z(t-1) is the cumulative state trend value at time t-1; ΔE(t) is the error change at time t; t k is the starting point of the kth time window; t k+1 is the end point of the kth time window; k is the index variable of the time window; E(t) is the total error at time t; Z(t) is the cumulative state trend value at time t; Z(t-1) is the cumulative state trend value at time t-1.

8. An intelligent method for evaluating the effect of immune cell therapy according to claim 7, characterized in that: The S2 specifically includes: By performing trend analysis on the stability index of the time window, the time window of abnormal state is identified, and the stability index analysis results of all time windows are summarized to generate a comprehensive treatment evaluation report.

9. An intelligent system for evaluating the effect of immune cell therapy, applied to the intelligent method for evaluating the effect of immune cell therapy as claimed in claim 1, characterized in that: Includes the following parts: Data collection and preprocessing module, adaptive fluctuation weight module, dynamic regular information entropy module, cumulative state trend module, total error and error change module, stability index module; The data collection and preprocessing module collects and preprocesses the original immune index data to obtain and output the immune index data; the data collection and preprocessing module is connected to the adaptive fluctuation weight module, the dynamic regular information entropy module, and the total error and error change module by means of data transmission; The adaptive fluctuation weight module receives the immune index data, calculates the adaptive fluctuation weight and outputs it; The adaptive fluctuation weight module is connected to the dynamic regular information entropy module through data transmission; The dynamic regular information entropy module receives the adaptive fluctuation weight and the immune index data, calculates the dynamic regular information entropy and outputs it; The dynamic regular information entropy module is connected to the cumulative state trend module through data transmission; The cumulative state trend module receives the dynamic regular information entropy, recursively updates the dynamic regular information entropy at each time, constructs a state trend that changes over time, obtains a cumulative state trend value and outputs it; The cumulative state trend module is connected with the total error and error change module and the stability index module through data transmission; The total error and error change module receives the cumulative state trend value and the immune index data, and calculates the total error and the error change; Based on the error change, evaluate the immediate treatment effect and generate a monitoring log; output the total error and error change; The total error and error change module is connected to the stability index module by means of data transmission; The stability index module receives the cumulative state trend value, the total error, and the error change, divides the treatment process into time windows, and calculates the stability index of the time window based on the total error, the error change, and the cumulative state trend value in the time window; by performing trend analysis on the stability index of the time window, the time window of abnormal state is identified, the stability index analysis results of all time windows are summarized, and a comprehensive treatment evaluation report is generated.

Images