A method and system for dynamically evaluating the efficacy of a hematological tumor-targeting drug

Abstract

The present application relates to the technical field of drug evaluation, and discloses a kind of blood tumor targeted drug curative effect dynamic evaluation method and system.The method comprises: by obtaining the feedback parameter containing action index and physiological index after targeted drug administration, and the difference calculation of feedback parameter and the benchmark feedback parameter under the condition of not being administered to output deviation parameter, and then generate drug efficacy fluctuation signal or risk early warning signal;And generate curative effect evaluation report based on drug efficacy fluctuation signal or risk early warning signal.The present application realizes the dynamic quantitative evaluation of curative effect by establishing the benchmark control of patient itself, objectively reflects the net effect of drug on patient, provides reliable individualized data basis for curative effect evaluation, improves the objectivity and accuracy of evaluation;And by filtering out random noise and transient fluctuation in data, ensure that the identified change starting point has high confidence, improve the accuracy and reliability of key event capture in dynamic evaluation process.

Description

Technical Field

[0001] This invention belongs to the field of drug evaluation technology, specifically relating to a method and system for dynamic evaluation of the efficacy of targeted drugs for hematological malignancies. Background Technology

[0002] With the development of precision medicine technology, targeted therapy against specific molecular targets has become one of the important treatment methods for conquering hematological malignancies such as leukemia, lymphoma, and multiple myeloma in modern cancer treatment. Compared with traditional chemotherapy, targeted drugs act specifically on specific signaling pathways of tumor cells, which can cause little or no damage to normal tissue cells, while inhibiting or killing tumor cells, thus improving the treatment effect and prognosis of patients.

[0003] Due to the heterogeneity of genotype, metabolic status and tumor microenvironment among patients, the efficacy of targeted drugs varies significantly among individuals. The same targeted drug may show different responses in different patients. Current clinical practice usually assesses the efficacy through macroscopic indicators after a period of medication. This method lacks early predictability and may cause patients to miss the best treatment window, and may also cause patients to bear unnecessary economic burden and drug toxicity. Furthermore, efficacy assessment methods such as regular hematological tests, imaging analyses, or bone marrow aspirations typically only reflect short-term treatment effects and are difficult to provide real-time dynamic feedback on the drug's effects at the cellular level. This results in a significant lag, making it difficult for doctors to adjust treatment plans specifically when adjusting dosages, combining medications, or changing drugs. Moreover, existing assessment indicators are unable to effectively distinguish between the cytotoxic side effects of drugs and their actual anti-tumor effects, and are also unable to identify secondary drug resistance in tumors at an early stage, increasing the uncertainty and risks in the treatment process.

[0004] In view of this, this application discloses a method and system for dynamic evaluation of the efficacy of targeted drugs for hematological malignancies. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for dynamic evaluation of the efficacy of targeted drugs for hematological malignancies. By analyzing patient feedback parameters in real time, the method can output the trend of drug efficacy changes and provide efficacy assessment or risk warning accordingly. Moreover, the data processing process will not cause any additional changes in the patient's physical signs.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A method for dynamic evaluation of the efficacy of targeted drugs for hematological malignancies, comprising the following steps: Obtain patient baseline feedback parameters and patient feedback parameters after administration of targeted drugs for hematological malignancies; When the feedback parameter deviates from the reference feedback parameter, a deviation parameter representing the deviation is determined; Generate efficacy evaluation reports based on deviation parameters; The feedback parameters include the efficacy and physiological indicators of hematologic malignancy targeted drugs; the baseline feedback parameters are the efficacy and physiological indicators obtained under conditions where hematologic malignancy targeted drugs have not been administered. Specifically, when the feedback parameter deviates from the baseline feedback parameter, the deviation parameter representing the deviation is determined by: obtaining the assessment time corresponding to the feedback parameter; constructing a function model based on the assessment time, administration plan, and feedback parameter, combined with the physiological difference function, to generate a sequence multi-sample output response, which includes the standard assessment output and the comparative assessment output; and determining the deviation parameter based on the sequence multi-sample output response.

[0007] Preferably, the generation of the efficacy evaluation report based on the deviation parameters includes: The deviation parameter is determined to be either positive or negative. When the deviation parameter is positive, a drug efficacy fluctuation signal is generated; when the deviation parameter is negative, a risk warning signal is generated; and a efficacy evaluation report is generated based on the drug efficacy fluctuation signal or the risk warning signal.

[0008] Preferably, the generation of the efficacy evaluation report based on the deviation parameters further includes: When the deviation parameter is a positive deviation and its value is greater than the preset deviation threshold for a duration greater than the preset time duration threshold, a drug resistance signal is generated and the drug resistance warning information is included in the efficacy evaluation report. When the deviation parameter is negative and its consecutive occurrences exceed a preset threshold, a side effect is identified. The corresponding time period is defined as the side effect time, and a risk assessment is performed based on the side effect time to generate a risk assessment result, which is then included in the efficacy assessment report.

[0009] Preferably, the generation of the efficacy evaluation report based on the deviation parameters further includes: When the frequency of adverse reactions is lower than a preset frequency threshold and the concentration of the distribution of the time points of occurrence is higher than a preset concentration threshold, the correlation between the adverse reactions and the targeted drugs for hematological malignancies is determined to be high, and the effectiveness confirmation information is included in the efficacy evaluation report. When the frequency of adverse reactions is not lower than a preset frequency threshold, or the concentration of the distribution of the time points of occurrence is not higher than a preset concentration threshold, the correlation between the adverse reactions and the targeted drugs for hematological malignancies is determined to be low, and the adverse reactions are excluded from the evaluation items.

[0010] Preferably, the method further includes: The difference between the standard evaluation output and the comparative evaluation output is calculated, and the difference is determined as the change in the drug's effect per unit time. The changes in the effect at each time point are sorted to determine the first time point where the fluctuation of the change in the effect exceeds the fluctuation threshold, and the first time point is determined as the start time of the change. Starting from the time of change, the time period is extended according to a preset time step to form a time period, and the corresponding feedback parameters within the time period are obtained to form a test sample. Multiple new change start times are generated based on the test samples. Based on the distribution density of the multiple new change start times, nodes with a distribution density greater than a preset density threshold are identified as change set nodes. The feedback parameters corresponding to the change set nodes are identified as central feedback parameters.

[0011] A dynamic evaluation of the efficacy of a targeted drug for hematological malignancies includes the following modules: The parameter acquisition module is used to acquire the patient's baseline feedback parameters and the patient's feedback parameters after administration of targeted drugs for hematological malignancies; The model building and response generation module is used to obtain the evaluation time corresponding to the feedback parameters; based on the evaluation time, administration plan and feedback parameters, a function model is constructed in combination with the physiological difference function to generate a multi-sample output response of the sequence. The deviation determination and central parameter identification module is used to determine the deviation parameters based on the output response of multiple samples in the sequence. The efficacy assessment and report generation module is used to respond to feedback parameters. When the feedback parameters deviate from the baseline feedback parameters, the deviation parameters that characterize the deviation are determined; and efficacy assessment reports are generated based on the deviation parameters.

[0012] Preferably, the feedback parameters include the efficacy indicators and physiological indicators of the hematologic malignancy targeted drug; the baseline feedback parameters are the efficacy indicators and physiological indicators obtained under the condition that the hematologic malignancy targeted drug has not been administered.

[0013] Preferably, the generation of the efficacy evaluation report based on the deviation parameters includes: The deviation parameter is determined to be either positive or negative. When the deviation parameter is positive, a drug efficacy fluctuation signal is generated; when the deviation parameter is negative, a risk warning signal is generated; and a efficacy evaluation report is generated based on the drug efficacy fluctuation signal or the risk warning signal.

[0014] Preferably, the generation of the efficacy evaluation report based on the deviation parameters further includes: When the deviation parameter is a positive deviation and its value is greater than the preset deviation threshold for a duration greater than the preset time duration threshold, a drug resistance signal is generated and the drug resistance warning information is included in the efficacy evaluation report. When the deviation parameter is negative and its consecutive occurrences exceed a preset threshold, a side effect is identified. The corresponding time period is defined as the side effect time, and a risk assessment is performed based on the side effect time to generate a risk assessment result, which is then included in the efficacy assessment report.

[0015] Preferably, the generation of the efficacy evaluation report based on the deviation parameters further includes: When the frequency of adverse reactions is lower than a preset frequency threshold and the concentration of the distribution of the time points of occurrence is higher than a preset concentration threshold, the correlation between the adverse reactions and the targeted drugs for hematological malignancies is determined to be high, and the effectiveness confirmation information is included in the efficacy evaluation report. When the frequency of adverse reactions is not lower than a preset frequency threshold, or the concentration of the distribution of the time points of occurrence is not higher than a preset concentration threshold, the correlation between the adverse reactions and the targeted drugs for hematological malignancies is determined to be low, and the adverse reactions are excluded from the evaluation items. Beneficial effects

[0016] This invention obtains feedback parameters, including efficacy and physiological indicators, after targeted drug administration. The difference between the feedback parameters and the baseline feedback parameters under no-drug conditions is calculated to output deviation parameters and generate drug efficacy fluctuation signals or risk warning signals. By establishing a patient-specific baseline control, dynamic quantitative assessment of efficacy can be achieved, reflecting the net effect of the drug on the patient, providing an individualized data basis for efficacy assessment, and improving the objectivity and accuracy of the assessment.

[0017] This invention outputs the response from multiple samples of the output sequence and determines the first fluctuation node by calculating the change in effect. It then determines the node of concentrated change based on the distribution density of the change initiation time. Thus, it accurately locates the key time point of drug efficacy or side effects from continuous feedback parameters, identifies the node with the most concentrated and significant changes, filters out random noise and transient fluctuations in the data, ensures that the identified change initiation point has high confidence, and improves the accuracy and reliability of key event capture in the dynamic evaluation process.

[0018] This invention distinguishes between positive deviations that reflect efficacy trends and negative deviations that reflect risk trends. It generates drug resistance signals by judging the duration of positive deviations, or determines the correlation between side effects and drugs by analyzing the frequency and distribution concentration of negative deviations. It outputs a multi-dimensional efficacy assessment report with clear clinical guidance, which can distinguish between the enhancement effect of drugs and side effects, as well as identify complex situations such as drug resistance and determine the causes of side effects. Attached Figure Description

[0019] Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is a system module diagram of the present invention. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments. It should be understood that the specific embodiments described herein are merely for explaining the invention and are not intended to limit the scope of protection of the invention. Example

[0021] Please refer to Figure 1 This embodiment provides a method for dynamic evaluation of the efficacy of targeted drugs for hematological malignancies, including the following steps: S1. Data Acquisition and Preparation: Acquire the drug administration regimens and treatment records of patients with hematologic malignancies; thereby establishing an initial dataset that includes drug interventions and individual patient background, providing a basis for personalized assessment and reducing assessment bias caused by individual differences. The administration protocol specifies the details of the drug intervention, including the drug name, dosage, administration method, and administration time; the patient's treatment record includes the patient's previous treatment records and basic information, including gender, age, disease course records, and physiological baselines such as weight, liver and kidney function baselines, or specific genotype information used for individualized calibration.

[0022] S2. Generation of feedback parameters and construction of response sequences; After the administration of the treatment plan, the patient's feedback parameters are obtained at multiple preset assessment time points; Among them, feedback parameters are direct evidence reflecting the effects and physiological influences of drugs in vivo. Specifically, they include: the concentration of specific tumor markers such as BCR-ABL fusion gene transcripts or the phosphorylation level of drug target kinases, which directly reflect drug efficacy; and physiological indicators reflecting the overall state of the body, such as complete blood count, liver and kidney function biochemical indicators, and electrocardiogram parameters. The acquired discrete feedback parameters, along with the corresponding evaluation time, dosage and administration time information in the administration protocol, are input into a pre-established calculation rule. This calculation rule includes a quantitative relationship between the administration information and the feedback parameters, as well as a series of correction coefficients associated with individual information such as the patient's age, gender, and physiological baseline. The correction coefficients are used to adjust the general response calculation parameters to specific calculation parameters for the current patient, reducing calculation noise introduced by individual physiological differences. By processing the data according to computational rules, structured time-series data is output, namely, the sequence multi-sample output response. This sequence multi-sample output response transforms discrete feedback parameters measured at different time points into continuous and standardized response curves that can characterize the patient's response trend, providing standardized data objects for dynamic analysis.

[0023] S3. Precise identification of change nodes and determination of central parameters; in order to accurately capture the key time points of drug action, it is necessary to analyze the output response of multiple samples of the sequence. The sequence multi-sample output response includes standard evaluation output and comparative evaluation output. The standard evaluation output is a predefined standard response curve, which is established based on statistical data of a large population or theoretical pharmacodynamic principles and represents the expected ideal response trend. The comparative evaluation output is the actual response curve generated based on the current patient's feedback parameters. The change in drug effect per unit time is obtained by calculating the difference between the standard assessment output and the comparative assessment output at each time point. The change in effect at each assessment time is compared with a preset fluctuation node threshold to determine the first time point when the change in effect first exceeds the fluctuation node threshold, and this first time point is defined as the first fluctuation node. The fluctuation node threshold is based on the ideal response curve of the drug, and can be taken as 0.8 × the clinical normal fluctuation limit. For example, if the normal fluctuation limit of the efficacy index of targeted drugs is ±5 ng / mL, then the preset fluctuation node threshold is ±4 ng / mL. The feedback parameter corresponding to the first fluctuation node is calibrated as the starting point of the change, and the time point when it occurs is set as the change start time. To confirm the stability of the change start time, a time window is set with the change start time as the center, and a more dense set of feedback parameters is obtained within this time window by increasing the sampling density. The set of feedback parameters is used to form a verification sample. The judgment logic for determining fluctuation nodes is applied again to the data points in the verification sample to obtain a set of candidate change time points. The distribution density of the candidate change time points is calculated, where the distribution density is a quantitative indicator for evaluating the density of candidate change time points on the time axis. The functional relationship for calculating the distribution density at candidate change time points is as follows: In the formula, This represents the distribution density, which means that at a given time point... The density of candidate change time points in the vicinity; the higher the value, the greater the likelihood that a key change occurred in the vicinity of that time. This represents any point in time, which is the target point in time used to calculate the density. This indicates the sample size, which represents the total number of candidate change time points; This represents the bandwidth, which is a smoothing parameter used to control the smoothness of density estimation and determines the range of the time neighborhood considered when calculating the density. The kernel function is defined as a non-negative function, a symmetric probability density function, and can be specifically chosen as a Gaussian kernel function. This is used to weight neighboring data points; Represents a set of candidate change time points, where, Indicates the first Candidate change time points.

[0024] Based on the distribution density of candidate change time points, nodes with a distribution density greater than a preset density threshold are identified as nodes in the change set, and the feedback parameters corresponding to the nodes in the change set are set as central feedback parameters; whereby the preset density threshold can be 0.6 when the Gaussian kernel function is selected for weighted calculation. Based on the central feedback parameters, a data collection time interval is traced back to ensure that key change information is fully captured. The data collection time interval is set to a time interval whose duration is greater than a preset time threshold. The preset time threshold can be the drug half-life or the clinical monitoring cycle. Specifically, if the drug half-life is 24 hours, the preset time threshold is 96 hours. To capture abrupt changes in response rate, multiple consecutive adjacent time periods can be set after drug administration. The output response corresponding to each of the multiple adjacent time periods can be extracted, and the difference between the output responses of each pair of adjacent time periods can be calculated. Furthermore, the rate of change of the difference between the output responses of each pair of adjacent time periods can be determined, and it can be judged whether the rate of change exceeds a preset rate of change threshold. If so, the starting time of the change when the rate of change exceeds the rate of change threshold is determined as the change start time. The preset rate of change threshold is based on the pharmacokinetic characteristics of the drug, or it can be selected as 15% / 24h according to the significant rate change in relevant standards.

[0025] S4. Benchmark comparison and deviation parameter quantification: The difference between the obtained feedback parameters and the benchmark feedback parameters is calculated to output the deviation parameters. Among them, the baseline feedback parameters are the effect indicators and physiological indicators obtained under similar physiological conditions without the administration of targeted drug therapy for hematological malignancies; specifically, the baseline feedback parameters can be the patient's own pre-treatment data or standardized baseline values ​​from healthy populations with similar characteristics. The difference between the feedback parameters and the baseline feedback parameters is calculated by comparing each parameter, resulting in a series of deviation parameters. When the value of the deviation parameter indicates a positive change in drug efficacy, such as a decrease in tumor markers, the deviation parameter is defined as a positive deviation; when its value indicates a potential negative impact, such as an increase in liver function indicators, it is defined as a negative deviation. By associating positive biases with positive identifiers and negative biases with negative identifiers, efficacy trends and risk trends are quantified in a structured manner. The efficacy trend is a judgment about the enhancement, weakening, or stabilization of drug efficacy derived by analyzing the changing patterns of continuous positive biases over a period of time. The risk trend is a judgment about the occurrence, persistence, or regression of adverse drug reactions derived by analyzing the changing patterns of continuous negative biases over a period of time.

[0026] S5. The generation of trend judgment and efficacy evaluation report is based on the quantified deviation parameters and executes the preset logical judgment rules to make trend judgment. Execute preset logical judgment rules to make trend judgments, including: generating indication information based on positive deviation to indicate that the drug is producing the expected therapeutic effect or other drug enhancement effects; and generating indication information based on negative deviation to indicate that the drug may cause unexpected physiological reactions or side effects or other adverse drug effects. When a positive deviation value or a type of positive deviation value is continuously greater than the preset deviation threshold and the duration exceeds the preset time duration threshold, it indicates that the phenomenon of tumor markers continuously rising after an initial decrease, indicating that tumor cells have developed resistance to the drug, is judged as a drug resistance signal, and drug resistance warning information is output as part of the efficacy evaluation report. When the number of consecutive occurrences of a negative deviation or a type of deviation exceeds a preset threshold, it is determined that the side reaction is persistent, and the corresponding time period is marked as the side reaction time. Based on factors such as the side reaction time and severity, a risk assessment is performed to generate a risk assessment result. The preset deviation threshold can be specifically referenced from the clinical efficacy evaluation standards of the drug, and is set to 20% of the baseline value of the tumor marker. Specifically, when the baseline value of the tumor marker is 10 ng / mL, the preset deviation threshold is 2 ng / mL. The preset duration threshold is determined according to the formation cycle of drug resistance, and in this embodiment, it is a duration of more than 10 days. The preset number threshold is set to 3 times according to the judgment requirements of persistent adverse events. To further distinguish the correlation between adverse reactions and drugs, the occurrence pattern of adverse reactions is analyzed. Specifically, when the frequency of adverse reactions is lower than a preset frequency threshold, and they are sporadic, and the distribution of their occurrence time points shows a high concentration after the drug administration time point, specifically when the concentration of the occurrence time distribution of adverse reactions is higher than a preset concentration threshold, the correlation between the adverse reaction and the hematologic malignancy targeted drug is determined to be high, and effectiveness confirmation information is output. Conversely, if the frequency of adverse reactions is not lower than the preset frequency threshold and is persistent or high-frequency, or if the concentration of their occurrence time distribution is not higher than the preset concentration threshold, the correlation between the adverse reaction and the targeted drug for hematological malignancies is determined to be low, and the adverse reaction is excluded from the evaluation items of this assessment. The preset frequency threshold is determined based on the drug administration cycle and the criteria for judging high-frequency adverse reactions. In this embodiment, the value is 3 times / 28 days. The preset concentration threshold is obtained based on the concentration threshold of similar drugs. In this embodiment, the value is 0.8. The concentration of the time distribution of adverse reactions is a numerical value used to quantify the degree of correlation between the time of occurrence of adverse reactions and the time of drug administration. Its calculation function includes: In the formula, This indicates the time points at which a group of adverse reactions occurred. This indicates the total number of adverse reaction events observed. Indicates the time of occurrence of adverse reactions, meaning the time point in a set of adverse reaction occurrences. The specific time point at which each adverse reaction event occurred; This represents a set of drug administration time points. This indicates the total number of doses administered, which means the total number of doses administered within the observation period; Indicates the time of administration, meaning the time point in a set of administration time points. The time point at which each drug administration event occurred; This indicates the search should be conducted no later than the time of adverse reaction occurrence. The most recent time of administration; Indicates a time delay, meaning the first... The time difference between an adverse reaction event and its most recent previous dosing event; This represents the set of calculated time delays, with the number of elements within the set matching the total number of observed adverse events. The variance of the set representing time delay; This is a concentration index used to quantify the degree of clustering of adverse reaction occurrence times relative to administration time. The closer the value is to 1, the more likely the adverse reaction occurs within a relatively fixed time interval after administration, indicating a high correlation; the closer the value is to 0, the more likely the adverse reaction occurs at different times, indicating a low correlation. S6. Drug efficacy fluctuation and risk warning: When the deviation parameter is positive and its fluctuation amplitude exceeds the preset fluctuation deviation threshold within the set time interval, output drug efficacy fluctuation signal; adjust the assessment level or prompt content in the efficacy assessment report accordingly based on the degree to which the fluctuation amplitude exceeds the preset fluctuation threshold. The fluctuation range can be obtained by calculating the standard deviation, range, or coefficient of variation of the deviation parameter within the time interval; the preset fluctuation deviation threshold is set at 50% of the mean of the positive deviation. When the deviation parameter is negative, a risk warning signal is generated to alert users to potential medication risks. The generated instructions, prompts, signals, and assessment results are integrated and compiled to produce a comprehensive and dynamic efficacy assessment report, which provides data support for clinical decision-making. Example

[0027] Please refer to Figure 2 This embodiment provides a dynamic evaluation system for the efficacy of targeted drugs for hematological malignancies, including the following modules: The parameter acquisition module is used to acquire feedback parameters of patients after administration of targeted drugs for hematological malignancies. It automatically or manually collects the required data by connecting to the data interface of the HIS medical information system, LIS laboratory information system or wearable health monitoring device. The feedback parameters include efficacy indicators and physiological indicators. Efficacy indicators can be data such as tumor marker levels, specific gene mutation abundance, or tumor size assessed by imaging, which directly reflect the drug's effect. Physiological indicators can be data such as blood routine, liver and kidney function, or electrocardiogram, which reflect the patient's overall physiological state. In the absence of targeted therapies for hematologic malignancies in patients, similar efficacy and physiological indicators were obtained as baseline feedback parameters and used as a baseline for subsequent comparisons. The assessment time corresponding to each feedback parameter was also obtained, and the patient's administration regimen was recorded. The patient's administration regimen included drug name, dosage, frequency of administration, and administration cycle information.

[0028] The model building and response generation module is used to build a function model that reflects individualized responses based on the acquired data, in order to generate sequence multi-sample output responses for in-depth analysis; After receiving feedback parameters, evaluation time, and administration plan, a function model is constructed by combining a pre-set or adaptively learned physiological difference function. Specifically, a log-normal distribution function can be used as the physiological difference function to construct the function model, or a pre-established calculation rule can be used. This calculation rule contains a quantitative relationship between drug administration information and feedback parameters, as well as a series of correction coefficients associated with individual information such as patient age, gender, and physiological baseline. Among them, the physiological difference function is used to quantify the inherent differences among different individuals in drug metabolism, distribution, and clearance. The functional relationship of the function model is as follows: In the formula, This represents the distribution density, which means that at a given time point... The density of candidate change time points in the vicinity; the higher the value, the greater the likelihood that a key change occurred in the vicinity of that time. This represents any point in time, which is the target point in time used to calculate the density. This indicates the sample size, which represents the total number of candidate change time points; This represents the bandwidth, which is a smoothing parameter used to control the smoothness of density estimation and determines the range of the time neighborhood considered when calculating the density. The kernel function is defined as a non-negative function, a symmetric probability density function, and can be specifically chosen as a Gaussian kernel function. This is used to weight neighboring data points; Represents a set of candidate change time points, where, Indicates the first One candidate time point of change; The function model processes the input data to generate a sequence multi-sample output response. Specifically, the sequence multi-sample output response includes a standard evaluation output and a comparative evaluation output. The standard evaluation output is the parameter change trajectory based on the model prediction under ideal drug efficacy, while the comparative evaluation output is the actual change trajectory generated based on the patient's actual feedback parameters.

[0029] The deviation determination and central parameter identification module is used to receive the output response of multiple samples of the sequence and perform in-depth analysis to determine key parameters; it calculates the difference between the standard evaluation output and the comparative evaluation output, and determines the difference as the change in the effect of the drug per unit time. The change in effect is used to reflect the degree of deviation between the actual drug effect and the expected drug effect. The changes in efficacy at each time point are sorted to determine the first time point when the change in efficacy first exceeds the fluctuation threshold. This time point is defined as the first fluctuation node, and the time point when it occurs is set as the change start time. The change start time marks the starting point where the efficacy may change significantly. Starting from the time of change, a time period is formed by extending the time step by a preset time step of one day or one medication cycle, and the corresponding feedback parameters within the time period are obtained to form a validation sample. Further, a set of candidate change time points were obtained based on the validation samples, and the distribution density of the candidate change time points was analyzed to accurately locate the core influencing factors. When the distribution density of certain nodes was found to be greater than the preset density threshold, these nodes were identified as change concentration nodes, and the feedback parameters corresponding to the change concentration nodes were identified as central feedback parameters. The central feedback parameters are considered to be the key driving factors leading to changes in drug efficacy. The deviation parameter, which characterizes the deviation, is determined by comparing the feedback parameter or the multi-sample output response of the sequence generated based on it with the benchmark feedback parameter.

[0030] The efficacy assessment and report generation module is used to integrate the analysis results and generate efficacy assessment reports. It receives deviation parameters determined by the deviation determination and central parameter identification module and determines whether the deviation parameters are positive or negative. Positive deviation usually indicates that the drug has produced the expected positive efficacy, while negative deviation usually indicates that an unexpected adverse physiological reaction has occurred. When the deviation parameter is negative, a risk warning signal is generated; when the deviation parameter is positive and its fluctuation amplitude within a preset time interval exceeds a preset fluctuation deviation threshold, a drug efficacy fluctuation signal is generated. When a deviation parameter is detected to be positive and its value is greater than the preset deviation threshold for a duration greater than the preset time duration threshold, it is determined that drug resistance may occur. At this time, a drug resistance signal is generated, and drug resistance warning information containing specific indicators and changing trends is included in the efficacy evaluation report. When a negative deviation parameter is detected and its consecutive occurrences exceed a preset threshold, a side effect is identified, and the corresponding time period is defined as the side effect time. Based on the side effect time, a risk assessment is further performed to evaluate the severity of the side effect, its impact on organ function, etc., and a risk assessment result is generated.

[0031] The correlation between the identified adverse reactions and the drug was also analyzed to improve the accuracy of the assessment. When the frequency of the adverse reaction was lower than the preset frequency threshold, it was considered to be an occasional event. When the concentration of the distribution of the time points of the adverse reaction was higher than the preset concentration threshold, it was considered to be strongly correlated with the time of occurrence and the treatment cycle. The correlation between the adverse reaction and the hematologic malignancy targeted drug was determined to be high, and this judgment was included in the report as an effectiveness confirmation. Conversely, if the frequency of occurrence is not lower than the preset frequency threshold, or the distribution concentration is not higher than the preset concentration threshold, the correlation is determined to be low, and the side effects can be excluded from this evaluation project to avoid interference from other complications or non-drug factors.

[0032] The generated drug efficacy fluctuation signals, risk warning signals, drug resistance indications, risk assessment results, and efficacy confirmation information are integrated into a structured and visualized efficacy assessment report, which is then presented to doctors or patients.

[0033] The above embodiments are only for illustrating the technical concept and features of the present invention, and are intended to enable those skilled in the art to understand and implement the present invention. They should not be construed as limiting the scope of protection of the present invention. All equivalent changes or modifications made in accordance with the spirit and essence of the present invention should be covered within the scope of protection of the present invention.

Claims

1. A method for dynamic evaluation of the efficacy of targeted drugs for hematological malignancies, characterized in that, Includes the following steps: Obtain patient baseline feedback parameters and patient feedback parameters after administration of targeted drugs for hematological malignancies; When the feedback parameter deviates from the reference feedback parameter, a deviation parameter representing the deviation is determined; Generate efficacy evaluation reports based on deviation parameters; The feedback parameters include the efficacy and physiological indicators of hematologic malignancy targeted drugs; the baseline feedback parameters are the efficacy and physiological indicators obtained under conditions where hematologic malignancy targeted drugs have not been administered. Specifically, when the feedback parameter deviates from the baseline feedback parameter, the deviation parameter representing the deviation is determined by: obtaining the assessment time corresponding to the feedback parameter; constructing a function model based on the assessment time, administration plan, and feedback parameter, combined with the physiological difference function, to generate a sequence multi-sample output response, which includes the standard assessment output and the comparative assessment output; and determining the deviation parameter based on the sequence multi-sample output response.

2. The method for dynamic evaluation of the efficacy of a targeted drug for hematological malignancies according to claim 1, characterized in that, The generation of efficacy evaluation reports based on deviation parameters includes: The deviation parameter is determined to be either positive or negative. When the deviation parameter is positive, a drug efficacy fluctuation signal is generated; when the deviation parameter is negative, a risk warning signal is generated; and a efficacy evaluation report is generated based on the drug efficacy fluctuation signal or the risk warning signal.

3. The method for dynamic evaluation of the efficacy of a targeted drug for hematological malignancies according to claim 2, characterized in that, The method for generating efficacy evaluation reports based on deviation parameters also includes: When the deviation parameter is a positive deviation and its value is greater than the preset deviation threshold for a duration greater than the preset time duration threshold, a drug resistance signal is generated and the drug resistance warning information is included in the efficacy evaluation report. When the deviation parameter is negative and its consecutive occurrences exceed a preset threshold, a side effect is identified. The corresponding time period is defined as the side effect time, and a risk assessment is performed based on the side effect time to generate a risk assessment result, which is then included in the efficacy assessment report.

4. The method for dynamic evaluation of the efficacy of a targeted drug for hematological malignancies according to claim 3, characterized in that, The method for generating efficacy evaluation reports based on deviation parameters also includes: When the frequency of adverse reactions is lower than a preset frequency threshold and the concentration of the distribution of the time points of occurrence is higher than a preset concentration threshold, the correlation between the adverse reactions and the targeted drugs for hematological malignancies is determined to be high, and the effectiveness confirmation information is included in the efficacy evaluation report. When the frequency of adverse reactions is not lower than a preset frequency threshold, or the concentration of the distribution of the time points of occurrence is not higher than a preset concentration threshold, the correlation between the adverse reaction and the targeted drug for hematological malignancies is determined to be low, and the adverse reaction is excluded from the evaluation item.

5. The method for dynamic evaluation of the efficacy of a targeted drug for hematological malignancies according to claim 1, characterized in that, The method further includes: The difference between the standard evaluation output and the comparative evaluation output is calculated, and the difference is determined as the change in the drug's effect per unit time. The changes in the effect at each time point are sorted to determine the first time point where the fluctuation of the change in the effect exceeds the fluctuation threshold, and the first time point is determined as the start time of the change. Starting from the time of change, the time period is extended according to a preset time step to form a time period, and the corresponding feedback parameters within the time period are obtained to form a test sample. Multiple new change start times are generated based on the test samples. Based on the distribution density of the multiple new change start times, nodes with a distribution density greater than a preset density threshold are identified as change set nodes. The feedback parameters corresponding to the change set nodes are identified as central feedback parameters.

6. A dynamic efficacy evaluation system for targeted drugs for hematological malignancies, characterized in that, Includes the following modules: The parameter acquisition module is used to acquire the patient's baseline feedback parameters and the patient's feedback parameters after administration of targeted drugs for hematological malignancies; The model building and response generation module is used to obtain the evaluation time corresponding to the feedback parameters; Based on the assessment time, administration plan and feedback parameters, a function model is constructed in combination with the physiological difference function to generate a sequence multi-sample output response; The deviation determination and central parameter identification module is used to determine the deviation parameters based on the output response of multiple samples in the sequence. The efficacy assessment and report generation module is used to respond to feedback parameters and determine the deviation parameters that characterize the deviation when the feedback parameters deviate from the baseline feedback parameters. Generate efficacy evaluation reports based on deviation parameters.

7. The dynamic efficacy evaluation system for targeted drugs for hematological malignancies according to claim 6, characterized in that, The feedback parameters include the efficacy and physiological indicators of hematologic malignancy targeted drugs; the baseline feedback parameters are the efficacy and physiological indicators obtained under conditions where hematologic malignancy targeted drugs have not been administered.

8. The dynamic efficacy evaluation system for targeted drugs for hematological malignancies according to claim 6, characterized in that, The generation of efficacy evaluation reports based on deviation parameters includes: The deviation parameter is determined to be either positive or negative. When the deviation parameter is positive, a drug efficacy fluctuation signal is generated; when the deviation parameter is negative, a risk warning signal is generated; and a efficacy evaluation report is generated based on the drug efficacy fluctuation signal or the risk warning signal.

9. The dynamic efficacy evaluation system for targeted drugs for hematological malignancies according to claim 8, characterized in that, The method for generating efficacy evaluation reports based on deviation parameters also includes: When the deviation parameter is a positive deviation and its value is greater than the preset deviation threshold for a duration greater than the preset time duration threshold, a drug resistance signal is generated and the drug resistance warning information is included in the efficacy evaluation report. When the deviation parameter is negative and its consecutive occurrences exceed a preset threshold, a side effect is identified. The corresponding time period is defined as the side effect time, and a risk assessment is performed based on the side effect time to generate a risk assessment result, which is then included in the efficacy assessment report.

10. The dynamic efficacy evaluation system for targeted drugs for hematological malignancies according to claim 9, characterized in that, The method for generating efficacy evaluation reports based on deviation parameters also includes: When the frequency of adverse reactions is lower than a preset frequency threshold and the concentration of the distribution of the time points of occurrence is higher than a preset concentration threshold, the correlation between the adverse reactions and the targeted drugs for hematological malignancies is determined to be high, and the effectiveness confirmation information is included in the efficacy evaluation report. When the frequency of adverse reactions is not lower than a preset frequency threshold, or the concentration of the distribution of the time points of occurrence is not higher than a preset concentration threshold, the correlation between the adverse reaction and the targeted drug for hematological malignancies is determined to be low, and the adverse reaction is excluded from the evaluation item.

Images