Method for evaluating curative effect of medicine for treating juvenile depression

Abstract

The invention belongs to the technical field of curative effect evaluation of teenager depression drugs, and particularly relates to a curative effect evaluation method of a drug for treating teenager depression. According to the method, through dynamic fusion analysis of multi-dimensional physiological and behavior data, the dynamic effect of drug intervention can be reflected more comprehensively and accurately, meanwhile, the data dimension is effectively reduced through dimension reduction processing and key feature vector extraction, capture of key information is ensured, and the accuracy of drug intervention is improved by constructing a drug use window and conducting quantitative scoring on the drug curative effect in the window. According to the method, evaluation of the drug curative effect and introduction of an interpolation compensation technology are realized, the problem of data missing is further solved, a continuous dynamic curative effect evaluation sequence is formed, and a time sequence evolution rule of the drug curative effect can be intuitively reflected according to a curative effect change trend identified by the dynamic curative effect evaluation sequence; the comprehensive curative effect stability index is calculated by combining the duration and the alternating frequency of the curative effect change trend in the medication cycle, and a scientific basis is provided for determining the drug curative effect grade.

Description

Technical Field

[0001] This invention belongs to the field of efficacy evaluation technology for drugs for treating adolescent depression, and specifically relates to a method for evaluating the efficacy of drugs for treating adolescent depression. Background Technology

[0002] With the increasing incidence of adolescent depression year by year, people have realized that traditional efficacy assessment methods have problems such as strong subjectivity, long cycle and delayed feedback, making it difficult to reflect the effect of drug intervention in a timely and accurate manner. However, with the development of information technology, the deep integration of wearable devices and big data analysis has provided technical possibilities for real-time dynamic monitoring of patients' physiological and behavioral characteristics, thereby enabling more timely and accurate capture of drug efficacy, allowing for objective quantitative assessment of the treatment effect of adolescent depression, and significantly improving the accuracy of clinical intervention.

[0003] While existing technologies offer some methods for evaluating drug efficacy based on physiological signals and behavioral data, they are largely limited to static analysis of single-dimensional data. This makes it difficult to comprehensively reflect the dynamic effects of drug intervention. Furthermore, they lack the ability to identify the evolutionary patterns of efficacy trends and cannot accurately reflect the temporal changes in patients' responses to drugs. Consequently, the evaluation results are unstable and cannot effectively support clinical decisions to adjust treatment plans. Consequently, it is difficult to dynamically track drug efficacy, thus limiting the optimization of individualized treatment strategies. Therefore, this invention proposes a method for evaluating the efficacy of drugs for treating adolescent depression to address the aforementioned problems. Summary of the Invention

[0004] The purpose of this invention is to provide a method for evaluating the efficacy of drugs for treating adolescent depression. This method can accurately identify the temporal evolution of efficacy trends by combining dynamic fusion analysis of multi-dimensional physiological and behavioral data, thereby achieving an effective evaluation of the stability of drug efficacy.

[0005] The specific technical solution adopted by this invention is as follows: A method for evaluating the efficacy of a drug for treating adolescent depression, comprising: Acquire physiological and behavioral data from patients with depression. The physiological data includes heart rate variability and skin conductance, while the behavioral data includes frequency of social activities and sleep quality scores. Physiological and behavioral data are normalized, and the normalized data is then dimensionality-reduced to extract key feature vectors. Obtain the medication window corresponding to each key feature vector, and quantitatively evaluate the drug efficacy within the medication window to obtain a drug efficacy score; The medication windows are arranged in chronological order on the medication timeline, and the missing time periods between adjacent medication windows are interpolated to form a dynamic efficacy evaluation sequence. The efficacy change trend is then identified based on the dynamic efficacy evaluation sequence, which includes efficacy improvement trend, efficacy fluctuation trend, and efficacy decline trend. Based on the duration and frequency of alternation of the upward, fluctuating, and downward trends during the medication cycle, a comprehensive efficacy stability index is calculated and output. The comprehensive efficacy stability index is then compared with the preset clinical efficacy standards to determine the efficacy level of the drug.

[0006] In a preferred embodiment, the step of obtaining physiological and behavioral data of patients with depression includes: Obtain the medication start time point for patients with depression, and construct a medication cycle based on the medication start time point; During the medication period, wearable devices were used to continuously collect data on patients' heart rate variability and skin conductance, which were then compiled into physiological indicator data for patients with depression. The frequency of patients' use of social applications, outdoor activity trajectories and sleep monitoring data were recorded by mobile terminals, and scores were obtained by combining the Pittsburgh Sleep Quality Index. The frequency of social activities and sleep quality scores are aggregated into behavioral data for patients with depression, and the behavioral data and physiological indicators are synchronously stored in a pre-defined data storage dataset.

[0007] In a preferred embodiment, the steps of normalizing physiological indicator data and behavioral data, and then performing dimensionality reduction on the normalized data to extract key feature vectors include: Physiological and behavioral data were aligned according to time series, and the min-max normalization method was used to eliminate dimensional differences. Covariance analysis was performed on the normalized physiological and behavioral data to extract principal component features and reduce the dimensionality to a two-dimensional feature space, resulting in low-dimensional feature vectors. Based on the temporal continuity of low-dimensional feature vectors, the Euclidean distance between adjacent time points is calculated using a sliding window mechanism, feature points that meet the distance threshold are selected, and a dynamic trajectory curve is constructed. The curvature change points of the trajectory curve are identified as key turning points, and feature intervals are defined with the key turning points as the center. Then, the curvature, slope and change amplitude are extracted as key feature vectors within the feature intervals.

[0008] In a preferred embodiment, the step of obtaining the medication window corresponding to each key feature vector and quantitatively evaluating the drug efficacy within the medication window to obtain a drug efficacy score includes: Obtain the time nodes corresponding to each key feature vector, map them to the medication cycle, and determine the corresponding medication window interval within the medication cycle. Each medication window is extended by a preset duration before and after the key feature vector time node. The average value of key feature vectors within the medication window is statistically analyzed and recorded as an indicator for efficacy evaluation. The efficacy assessment indicators are compared with the efficacy benchmark values, the deviation between the assessment indicators and the efficacy benchmark values ​​is calculated, and the deviation is then mapped to a preset scoring table to obtain the corresponding drug efficacy score.

[0009] In a preferred embodiment, the step of interpolating and compensating for missing time periods between adjacent medication windows to form a dynamic efficacy evaluation sequence includes: Identify the blank time intervals between adjacent medication windows and calculate the duration of the blank time intervals; Obtain the drug efficacy scores within the two medication windows before and after the blank time interval, input them into the preset efficacy change function, and record the output of the efficacy change function as the interpolation compensation score. Discrete sampling is performed on the blank time interval using a preset time dimension, and interpolation compensation scores are assigned to each sampling time point to generate an interpolation compensation score sequence. The interpolated compensation score sequence is spliced ​​with the score sequences under the medication windows before and after the blank time interval to form a continuous dynamic efficacy assessment sequence.

[0010] In a preferred embodiment, the step of identifying the trend of efficacy changes based on the dynamic efficacy assessment sequence includes: The dynamic efficacy assessment sequence is segmented over time, and the average rate of change of efficacy score over time is calculated within each medication window as the baseline unit. When the average rate continues to increase positively and exceeds N medication windows consecutively, it indicates that the drug inhibition is effective, and the change in efficacy within the medication window is marked as an upward trend in efficacy. When the average rate changes negatively continuously and exceeds N medication windows, it is determined that the efficacy has weakened or the condition has rebounded, and the change in efficacy within the medication window is marked as a downward trend in efficacy. If the average rate shows both positive growth and negative changes within less than N medication windows, the drug efficacy is considered unstable, and the efficacy changes within the medication windows are not marked as efficacy fluctuation trends.

[0011] In a preferred embodiment, the step of calculating and outputting the comprehensive efficacy stability index based on the duration and alternation frequency of the upward trend, fluctuation trend, and downward trend during the medication cycle includes: The total duration of the efficacy improvement trend, the total duration of the efficacy fluctuation trend, and the total duration of the efficacy decline trend were statistically analyzed within the medication cycle. Calculate the proportion of the total duration of the efficacy improvement trend, the total duration of the efficacy fluctuation trend, and the total duration of the efficacy decline trend within the medication cycle, and record them as the first proportion coefficient, the second proportion coefficient, and the third proportion coefficient, respectively. Extract the frequency of alternation between the trends of therapeutic efficacy improvement, fluctuation, and decline, and record it as the trend alternation frequency coefficient; Obtain the time point distribution of the efficacy improvement trend, efficacy fluctuation trend, and efficacy decline trend within the medication cycle, and match the time sequence weight coefficients of the efficacy improvement trend, efficacy fluctuation trend, and efficacy decline trend within the medication cycle based on the time point distribution. The first proportion coefficient, the second proportion coefficient, the third proportion coefficient, and the trend alternation frequency coefficient are weighted and summed, and combined with the time series weight coefficient for nonlinear mapping to generate the comprehensive efficacy stability index.

[0012] In a preferred embodiment, the step of comparing the comprehensive efficacy stability index with a preset clinical efficacy standard to determine the efficacy level of the drug includes: Obtain the clinical efficacy grading intervals and compare the comprehensive efficacy stability index with the clinical efficacy grading intervals; If the overall efficacy stability index exceeds the upper limit of the clinical efficacy grading range, the drug efficacy level is determined to be significantly improved. If the overall efficacy stability index is within the clinical efficacy grading range, the drug efficacy level is determined to be effective. If the overall efficacy stability index is lower than the lower limit of the clinical efficacy grading range, the drug efficacy level is determined to be ineffective, and an early warning is sent to the patient monitoring terminal.

[0013] The present invention also provides an efficacy evaluation system for a drug for treating adolescent depression, using the above-mentioned efficacy evaluation method for a drug for treating adolescent depression, comprising: The data acquisition module is used to acquire physiological and behavioral data of patients with depression. The physiological data includes heart rate variability and skin conductance response, and the behavioral data includes the frequency of social activities and sleep quality scores. The data processing module is used to normalize physiological indicator data and behavioral data, and to perform dimensionality reduction on the normalized data to extract key feature vectors. The quantitative evaluation module is used to obtain the medication window corresponding to each key feature vector, and to quantitatively evaluate the efficacy of the drugs within the medication window to obtain a drug efficacy score. The trend recognition module is used to arrange the medication windows in the medication timeline according to the occurrence sequence, and to interpolate and compensate for the missing time periods between adjacent medication windows to form a dynamic efficacy evaluation sequence. Then, the efficacy change trend is identified based on the dynamic efficacy evaluation sequence, including the efficacy change trend, the improvement trend, the fluctuation trend, and the decline trend. The efficacy evaluation module is used to calculate and output a comprehensive efficacy stability index based on the duration and frequency of alternation of the upward trend, fluctuation trend and downward trend during the medication cycle. The comprehensive efficacy stability index is then compared with the preset clinical efficacy standards to determine the efficacy level of the drug.

[0014] And, an electronic device, the electronic device comprising: At least one processor; and a memory communicatively connected to the at least one processor; The memory stores a computer program that can be executed by the at least one processor, which enables the at least one processor to perform the above-described efficacy evaluation method for drugs used to treat adolescent depression.

[0015] The technical effects achieved by this invention are as follows: This invention, through dynamic fusion analysis of multi-dimensional physiological and behavioral data, can more comprehensively and accurately reflect the dynamic effects of drug intervention. Simultaneously, by utilizing dimensionality reduction and key feature vector extraction, it effectively reduces data dimensionality, ensuring the capture of crucial information. By constructing a medication window and quantifying the drug efficacy within that window, it achieves drug efficacy assessment. The introduction of interpolation compensation technology further compensates for data gaps, forming a continuous dynamic efficacy assessment sequence. Based on the efficacy change trend identified by the dynamic efficacy assessment sequence, it can intuitively reflect the temporal evolution of drug efficacy. The comprehensive efficacy stability index, calculated by combining the duration and frequency of efficacy change trends within the medication cycle, provides a scientific basis for determining the drug efficacy level. Comparison with preset clinical efficacy standards makes the determination of drug efficacy level more objective and accurate, helping clinicians to adjust treatment plans in a timely manner and optimize individualized treatment strategies. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the method flow of the present invention.

[0017] Figure 2 This is a schematic diagram of the system modules of the present invention.

[0018] Figure 3 This is a schematic diagram of the electronic device structure of the present invention. Detailed Implementation

[0019] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0020] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0021] Secondly, the term "an embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in a preferred embodiment" appearing in different places throughout this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that mutually excludes other embodiments.

[0022] With the accelerating pace of social life and increasing environmental pressure, depression has become an increasingly common mental health problem among adolescents. To ensure the effective maintenance of the physical and mental health of adolescents, a more comprehensive assessment mechanism needs to be established for the drug treatment of adolescent depression, so as to effectively improve the mental health of adolescents.

[0023] Please see Figure 1 As shown, the present invention provides a method for evaluating the efficacy of drugs for treating adolescent depression, comprising: S1. Obtain physiological and behavioral data of patients with depression. The physiological data includes heart rate variability and skin conductance response, and the behavioral data includes the frequency of social activities and sleep quality scores.

[0024] In step S1, when evaluating the efficacy of antidepressants, it is first necessary to collect multidimensional physiological and behavioral data from patients with depression. Physiological indicators, such as heart rate variability and skin conductance, can reflect the regulatory state of the patient's autonomic nervous system and are physiological markers for assessing mood fluctuations and stress levels. Behavioral data, such as the frequency of social activities and sleep quality scores, can directly reflect the degree of improvement in the patient's daily functional state. The combination of these two data can more comprehensively reflect the impact of drug intervention on the patient's psychophysiological state, providing corresponding data support for subsequent efficacy analysis. The steps for obtaining physiological and behavioral data from patients with depression include: Obtain the medication start time point for patients with depression, and construct a medication cycle based on the medication start time point; During the medication period, wearable devices were used to continuously collect data on patients' heart rate variability and skin conductance, which were then compiled into physiological indicator data for patients with depression. The frequency of patients' use of social applications, outdoor activity trajectories and sleep monitoring data were recorded by mobile terminals, and scores were obtained by combining the Pittsburgh Sleep Quality Index. The frequency of social activities and sleep quality scores are aggregated into behavioral data for patients with depression, and the behavioral data and physiological indicators are synchronously stored in a pre-defined data storage dataset.

[0025] Specifically, in acquiring physiological and behavioral data from adolescent patients with depression, the first step is to record the starting time of medication use. This starting time serves as a key benchmark for constructing the medication cycle. The medication cycle needs to be set based on the drug's half-life and clinical dosing regimen, and the specific settings will vary depending on the actual situation. During the medication cycle, patients continuously collect heart rate variability and skin conductance data by wearing wearable devices. Wearable devices include smart bracelets, smartwatches, and other devices with physiological monitoring functions, or dedicated physiological recorders with medical-grade sensors. Simultaneously, mobile terminals such as smartphones and tablets are used to record the frequency of social media use, outdoor activity trajectories, and sleep data collected by built-in sleep monitoring modules. These data are then combined with the Pittsburgh Sleep Quality Index (PSI) for comprehensive scoring, thereby obtaining social activity frequency and sleep quality scores. Finally, these social activity frequency and sleep quality scores are aggregated into behavioral data and stored together with the synchronously collected physiological data in a pre-defined data storage dataset, ensuring the time series alignment and integrity of the data.

[0026] S2. Normalize the physiological and behavioral data, and then reduce the dimensionality of the normalized data to extract key feature vectors.

[0027] In step S2, after the collection of the patient's physiological and behavioral data is completed, to ensure the comparability of data from different dimensions, the physiological and behavioral data will be normalized to unify the dimensions of different data. Then, the normalized data will be dimensionality reduced to extract key feature vectors, providing crucial feature input for subsequent efficacy evaluation. The steps of normalizing the physiological and behavioral data, reducing the dimensionality of the normalized data, and extracting key feature vectors include: Physiological and behavioral data were aligned according to time series, and the min-max normalization method was used to eliminate dimensional differences. Covariance analysis was performed on the normalized physiological and behavioral data to extract principal component features and reduce the dimensionality to a two-dimensional feature space, resulting in low-dimensional feature vectors. Based on the temporal continuity of low-dimensional feature vectors, the Euclidean distance between adjacent time points is calculated using a sliding window mechanism, feature points that meet the distance threshold are selected, and a dynamic trajectory curve is constructed. The curvature change points of the trajectory curve are identified as key turning points, and feature intervals are defined with the key turning points as the center. Then, the curvature, slope and change amplitude are extracted as key feature vectors within the feature intervals.

[0028] Specifically, in the process of extracting key feature vectors, physiological indicator data and behavioral data are first arranged and aligned chronologically to ensure data consistency over time. Then, the min-max normalization method is used to unify physiological indicator data and behavioral data of different dimensions to a specific numerical range, eliminating the interference of dimensional differences on subsequent analysis. Of course, normalization can also be performed using methods such as Z-score standardization. Subsequently, covariance analysis is performed on the normalized physiological indicator data and behavioral data to extract principal component features, and simultaneously reduce the dimensionality to a two-dimensional feature space. By taking advantage of the continuity of low-dimensional feature vectors over time, and combining the sliding window mechanism, the Euclidean distance between adjacent time points is calculated. Feature points that meet the conditions are selected according to the pre-set distance threshold. A dynamic trajectory curve is constructed using the feature points. Finally, the point where the curvature of the dynamic trajectory curve changes abruptly is identified as the key turning point. The feature interval is delineated with the key turning point as the center. Within the defined feature interval, parameters such as curvature, slope, and change amplitude are further extracted and then vectorized simultaneously to obtain the key feature vector for efficacy evaluation.

[0029] S3. Obtain the medication window corresponding to each key feature vector, and quantitatively evaluate the efficacy of the drug within the medication window to obtain the drug efficacy score;

[0030] In step S3, after the key feature vector is output, its corresponding time segment is located as the medication window. Then, the drug efficacy within each medication window is quantitatively evaluated based on the key feature vector, and this is reflected through a drug efficacy score. The steps of obtaining the medication window corresponding to each key feature vector and quantitatively evaluating the drug efficacy within the medication window to obtain the drug efficacy score include: Obtain the time nodes corresponding to each key feature vector, map them to the medication cycle, and determine the corresponding medication window interval within the medication cycle. Each medication window is extended by a preset duration before and after the key feature vector time node. The average value of key feature vectors within the medication window is statistically analyzed and recorded as an indicator for efficacy evaluation. The efficacy assessment indicators are compared with the efficacy benchmark values, the deviation between the assessment indicators and the efficacy benchmark values ​​is calculated, and the deviation is then mapped to a preset scoring table to obtain the corresponding drug efficacy score.

[0031] Specifically, when determining the drug efficacy score within a medication window, the process first obtains the time nodes corresponding to each key feature vector and maps these time nodes to the medication cycle. This clarifies the medication window corresponding to each key feature vector within the medication cycle. Each medication window is determined by extending a preset duration forward and backward from the key feature vector's time node. The average value of the key feature vectors within the medication window is then calculated and recorded as the efficacy evaluation index. This index is then compared to a pre-set efficacy benchmark value to calculate the deviation. The efficacy benchmark value is set based on the average feature vector value of patients in a stable state from historical clinical data. The deviation is calculated using the formula: Deviation = (Evaluation Index - Efficacy Benchmark Value) / Efficacy Benchmark Value. Finally, based on the deviation, it is mapped to a preset scoring table, thus obtaining and outputting the corresponding drug efficacy score. This provides a clear picture of the drug efficacy performance within each medication window, offering data support for doctors to optimize medication regimens.

[0032] S4. Arrange the medication windows on the medication timeline according to the order of occurrence, and interpolate to compensate for the missing time periods between adjacent medication windows to form a dynamic efficacy evaluation sequence. Then, identify the efficacy change trend based on the dynamic efficacy evaluation sequence. The efficacy change trend includes efficacy improvement trend, efficacy fluctuation trend, and efficacy decline trend.

[0033] In step S4, after the drug efficacy score is output, all medication windows are arranged according to their actual occurrence sequence on the medication timeline to construct a preliminary efficacy assessment time series. Considering the possibility of data loss in some time periods due to various factors during actual data collection, this embodiment uses an interpolation algorithm to compensate for missing time periods between adjacent medication windows to ensure the continuity and accuracy of efficacy assessment, thereby forming a complete and continuous dynamic efficacy assessment sequence. Then, based on this dynamic efficacy assessment sequence, the efficacy change trend contained therein is identified to reflect the evolution law of drug efficacy. The step of interpolating and compensating for missing time periods between adjacent medication windows to form a dynamic efficacy assessment sequence includes: Identify the blank time intervals between adjacent medication windows and calculate the duration of the blank time intervals; Obtain the drug efficacy scores within the two medication windows before and after the blank time interval, input them into the preset efficacy change function, and record the output of the efficacy change function as the interpolation compensation score. Discrete sampling is performed on the blank time interval using a preset time dimension, and interpolation compensation scores are assigned to each sampling time point to generate an interpolation compensation score sequence. The interpolated compensation score sequence is spliced ​​with the score sequences under the medication windows before and after the blank time interval to form a continuous dynamic efficacy assessment sequence.

[0034] Specifically, when outputting the dynamic efficacy evaluation sequence, the system first identifies the blank time intervals between adjacent medication windows and calculates the specific duration of these blank time intervals. Then, it obtains the drug efficacy scores for the two medication windows before and after the blank time interval. These scores are input into a pre-defined efficacy change function for calculation, resulting in the output of the corresponding interpolation compensation score. The expression for the efficacy change function is as follows: In the formula, f(t) represents the interpolation compensation score, y1 represents the drug efficacy score of the medication window before the blank time interval, y2 represents the drug efficacy score of the medication window before and after the blank time interval, t represents the sampling time point within the blank time interval, t1 represents the center time point of the medication window before the blank time interval, and t2 represents the center time point of the medication window after the blank time interval. The interpolation compensation score is used to fill in the missing data in the time interval. Then, the blank time interval is discretely sampled with a preset time granularity to ensure the uniformity and representativeness of the sampling. The interpolation compensation score is used to assign values ​​to each sampling time point, thereby generating the interpolation compensation score sequence. Finally, the generated interpolation compensation score sequence is spliced ​​with the score sequences under the medication windows before and after the blank time interval to form a continuous and complete dynamic efficacy evaluation sequence, providing a reliable data foundation for subsequent efficacy trend identification.

[0035] Secondly, the steps for identifying trends in efficacy based on dynamic efficacy assessment sequences include: The dynamic efficacy assessment sequence is segmented over time, and the average rate of change of efficacy score over time is calculated within each medication window as the baseline unit. When the average rate continues to increase positively and exceeds N medication windows consecutively, it indicates that the drug inhibition is effective, and the change in efficacy within the medication window is marked as an upward trend in efficacy. When the average rate changes negatively continuously and exceeds N medication windows, it is determined that the efficacy has weakened or the condition has rebounded, and the change in efficacy within the medication window is marked as a downward trend in efficacy. If the average rate shows both positive growth and negative changes within less than N medication windows, the drug efficacy is considered unstable, and the efficacy changes within the medication windows are not marked as efficacy fluctuation trends.

[0036] In the above process, when identifying trends in efficacy, the dynamic efficacy assessment sequence is first segmented chronologically, with each medication window serving as the basic unit of analysis. The dynamic characteristics of drug efficacy are quantified by calculating the average rate of change in efficacy scores over time within each medication window. Specifically, when the average rate shows a sustained positive increase, and this upward trend continues for more than N medication windows (N being a natural number greater than or equal to 3), the drug's inhibitory effect is deemed effective. The efficacy changes within the corresponding time period are then marked as an upward trend, indicating that the patient's psychological and physiological state is gradually improving. Conversely, if the average rate shows a sustained negative change, and this downward trend continues for more than [number missing] medication windows, the drug's inhibitory effect is considered effective. If the drug efficacy is reduced after N medication windows, it indicates that the drug's effectiveness may be weakening or that the patient's condition has rebounded. In this case, the efficacy changes within the corresponding time period will be marked as a declining efficacy trend, prompting the doctor to reassess and adjust the treatment plan. In addition, when the average rate is within a time range of less than N medication windows, and there are both positive increases and negative changes, the drug efficacy will be determined to be unstable during this period. The efficacy changes within the corresponding time period will be marked as an efficacy fluctuation trend. This situation may reflect individual differences in the patient's response to the drug or be affected by other external factors. In this case, the doctor's intervention is also required for comprehensive assessment and intervention to avoid the occurrence of disease deterioration or treatment failure.

[0037] S5. Based on the duration and frequency of alternation of the upward trend, fluctuation trend and downward trend during the medication cycle, calculate and output the comprehensive efficacy stability index, and then compare the comprehensive efficacy stability index with the preset clinical efficacy standard to determine the efficacy level of the drug.

[0038] In step S5, after the efficacy trend is output, the corresponding efficacy stability determination can be performed. This involves calculating a comprehensive efficacy stability index based on the duration and frequency of alternation of the upward, fluctuating, and downward trends within the medication cycle. The comprehensive efficacy stability index is then compared with preset clinical efficacy standards to determine the final efficacy level of the drug, providing doctors and patients with an intuitive efficacy assessment reference. The step of calculating and outputting the comprehensive efficacy stability index based on the duration and frequency of alternation of the upward, fluctuating, and downward trends within the medication cycle includes: The total duration of the efficacy improvement trend, the total duration of the efficacy fluctuation trend, and the total duration of the efficacy decline trend were statistically analyzed within the medication cycle. Calculate the proportion of the total duration of the efficacy improvement trend, the total duration of the efficacy fluctuation trend, and the total duration of the efficacy decline trend within the medication cycle, and record them as the first proportion coefficient, the second proportion coefficient, and the third proportion coefficient, respectively. Extract the frequency of alternation between the trends of therapeutic efficacy improvement, fluctuation, and decline, and record it as the trend alternation frequency coefficient; Obtain the time point distribution of the efficacy improvement trend, efficacy fluctuation trend, and efficacy decline trend within the medication cycle, and match the time sequence weight coefficients of the efficacy improvement trend, efficacy fluctuation trend, and efficacy decline trend within the medication cycle based on the time point distribution. The first proportion coefficient, the second proportion coefficient, the third proportion coefficient, and the trend alternation frequency coefficient are weighted and summed, and combined with the time series weight coefficient for nonlinear mapping to generate the comprehensive efficacy stability index.

[0039] Specifically, when calculating the comprehensive efficacy stability index, the total duration of the efficacy improvement trend, the total duration of the efficacy fluctuation trend, and the total duration of the efficacy decline trend are first counted within the medication cycle. This reflects the duration of different efficacy change trends throughout the overall medication cycle. Then, the proportion of each of these three total durations within the medication cycle is calculated and recorded as the first proportion coefficient, the second proportion coefficient, and the third proportion coefficient, respectively. This provides a clear understanding of the time proportion of each efficacy change trend within the medication cycle. Furthermore, the frequency of alternation between the efficacy improvement trend, fluctuation trend, and decline trend is extracted and recorded as the trend alternation frequency coefficient. The trend alternation frequency coefficient reflects the alternation of different efficacy change trends within the medication cycle. In addition to the frequency of occurrence, the distribution of efficacy improvement trends, efficacy fluctuation trends, and efficacy decline trends within the treatment cycle is also obtained. Because the impact of efficacy change trends at different time points on the overall efficacy assessment may vary, time-series weighting coefficients for efficacy improvement trends, efficacy fluctuation trends, and efficacy decline trends within the treatment cycle are matched based on their time-series distribution. Finally, the first, second, and third proportion coefficients are weighted and summed with the trend alternation frequency coefficient. By combining the duration proportion and alternation frequency information of different efficacy change trends, and then performing a non-linear mapping with the time-series weighting coefficients, a comprehensive efficacy stability index reflecting the efficacy stability of the drug throughout the entire treatment cycle can be generated. The formula for calculating the comprehensive efficacy stability index is as follows: In the formula, I represents the overall efficacy stability index, T represents the total duration of the medication cycle, and C... p C f and C d These represent the first, second, and third proportion coefficients, respectively. α, β, and χ represent the positive contribution weight when drug efficacy improves, the neutral impact weight when drug efficacy fluctuates, and the negative penalty weight when drug efficacy declines, respectively. W kThe time-series weighting coefficient is dynamically adjusted based on the start and end times of each trend segment in the time series. It reflects the differences in the impact of early, mid, and late-stage efficacy changes on overall stability. For example, early efficacy fluctuations may be considered normal due to the body's adaptation process, and their time-series weight is low. However, similar fluctuations occurring in the mid and late stages may indicate difficulties in maintaining efficacy, and are therefore given a higher penalty weight. δ represents the penalty weight for the frequency of trend alternation, F represents the frequency coefficient of trend alternation, and λ represents the nonlinear mapping index, which is used to adjust the sensitivity of the comprehensive efficacy stability index to the frequency of trend alternation. Its value is dynamically set according to the specific drug type and treatment cycle to ensure the applicability and discriminative power of the index in different clinical scenarios.

[0040] Secondly, the steps of comparing the comprehensive efficacy stability index with the pre-set clinical efficacy criteria to determine the efficacy level of the drug include: Obtain the clinical efficacy grading intervals and compare the comprehensive efficacy stability index with the clinical efficacy grading intervals; If the overall efficacy stability index exceeds the upper limit of the clinical efficacy grading range, the drug efficacy level is determined to be significantly improved. If the overall efficacy stability index is within the clinical efficacy grading range, the drug efficacy level is determined to be effective. If the overall efficacy stability index is lower than the lower limit of the clinical efficacy grading range, the drug efficacy level is determined to be ineffective, and an early warning is sent to the patient monitoring terminal.

[0041] In determining the efficacy level of antidepressants, the first step is to obtain the clinical efficacy grading interval under the clinical efficacy criteria. This interval is defined based on extensive clinical trial data and aims to clarify the quantitative range corresponding to different efficacy levels. Then, the calculated comprehensive efficacy stability index is compared with this clinical efficacy grading interval. If the comprehensive efficacy stability index exceeds the upper limit of the interval, it indicates that the drug has shown extremely stable and significant efficacy improvement throughout the entire treatment period, and the patient's depressive symptoms have been significantly relieved and improved. In this case, the efficacy level of the drug is determined to be significantly improved, indicating that the antidepressant performs well in the treatment of adolescent depression and can bring significant therapeutic effects to patients. If the comprehensive efficacy stability index is within the clinical efficacy grading interval... Within a certain period, it indicates that the medication has played a positive role in the treatment process, and the patient's depressive symptoms have shown a relatively obvious improvement trend, but have not yet reached an extremely significant level. At this time, the medication efficacy level is determined to be effective, indicating that the medication has a certain effect on the treatment of adolescent depression and can continue to be used in clinical treatment for further observation and optimization. If the comprehensive efficacy stability index is lower than the lower limit of the clinical efficacy grading interval, it indicates that the medication's efficacy performance during the treatment process is poor, the patient's depressive symptoms have not been effectively controlled, and may even have worsened. At this time, the medication efficacy level is determined to be ineffective, and an early warning will be sent to the patient's monitoring terminal to promptly remind doctors and the patient's family to pay attention to changes in the condition so as to adjust the treatment plan in a timely manner, avoid further deterioration of the condition, and ensure the health and safety of adolescent patients with depression.

[0042] Please see Figure 2 A system for evaluating the efficacy of a drug for treating adolescent depression, using the aforementioned method for evaluating the efficacy of a drug for treating adolescent depression, including: The data acquisition module is used to acquire physiological and behavioral data of patients with depression. The physiological data includes heart rate variability and skin conductance response, and the behavioral data includes the frequency of social activities and sleep quality scores. The data processing module is used to normalize physiological indicator data and behavioral data, and to perform dimensionality reduction on the normalized data to extract key feature vectors. The quantitative evaluation module is used to obtain the medication window corresponding to each key feature vector, and to quantitatively evaluate the efficacy of the drugs within the medication window to obtain a drug efficacy score. The trend recognition module is used to arrange the medication windows in the medication timeline according to the occurrence sequence, and to interpolate and compensate for the missing time periods between adjacent medication windows to form a dynamic efficacy evaluation sequence. Then, the efficacy change trend is identified based on the dynamic efficacy evaluation sequence, including the efficacy change trend, the improvement trend, the fluctuation trend, and the decline trend. The efficacy evaluation module is used to calculate and output a comprehensive efficacy stability index based on the duration and frequency of alternation of the upward trend, fluctuation trend and downward trend during the medication cycle. The comprehensive efficacy stability index is then compared with the preset clinical efficacy standards to determine the efficacy level of the drug.

[0043] The execution process of the above-mentioned efficacy assessment system corresponds to the steps of the aforementioned efficacy assessment method, and will not be repeated here.

[0044] Please see Figure 3 An electronic device, comprising: At least one processor; and memory that is communicatively connected to at least one processor; The memory stores a computer program that can be executed by at least one processor, which enables the at least one processor to perform the above-described efficacy evaluation method for drugs used to treat adolescent depression.

[0045] The processor of the aforementioned electronic device can be at least one central processing unit (CPU) or graphics processing unit (GPU) to execute computer program instructions stored in memory. The memory can take various forms, such as random access memory (RAM), read-only memory (ROM), or a combination of both, to persistently store computer programs and various types of data generated during operation. In addition, the electronic device may also include an arithmetic logic unit (ALU), input devices, and output devices. The ALU can be an arithmetic logic unit (ALU) responsible for performing basic arithmetic and logical operations. Input devices include a touch screen, microphone, or biosignal acquisition module to receive user operation instructions and physiological data. Output devices include a display screen, speaker, or vibration motor to realize visual feedback and early warning prompts of evaluation results.

[0046] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, apparatus, article, or method. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, apparatus, article, or method that includes that element.

[0047] The above description is merely a preferred embodiment of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention. Structures, devices, and operating methods not specifically described or explained in this invention are implemented according to conventional methods in the art unless otherwise specified or limited.

Claims

1. A method for evaluating the efficacy of a drug for treating depression in adolescents, characterized by: The method comprises the steps of: acquiring physiological index data and behavior data of a depression patient, wherein the physiological index data comprises heart rate variability and skin electric response, and the behavior data comprises social activity frequency and sleep quality score; normalizing the physiological index data and the behavior data, and performing dimension reduction processing on the normalized data to extract key feature vectors; acquiring a medication window corresponding to each key feature vector, and quantitatively evaluating the drug efficacy in the medication window to obtain a drug efficacy score; arranging the medication windows in a medication time axis according to the occurrence time sequence, interpolating and compensating for the missing period between adjacent medication windows, forming a dynamic efficacy evaluation sequence, and identifying the efficacy change trend according to the dynamic efficacy evaluation sequence, wherein the efficacy change trend comprises an efficacy improvement trend, an efficacy fluctuation trend and an efficacy decline trend; calculating and outputting a comprehensive efficacy stability index according to the duration and alternating frequency of the improvement trend, the fluctuation trend and the decline trend in the medication cycle, and comparing the comprehensive efficacy stability index with a preset clinical efficacy standard to determine the efficacy grade of the drug.

2. The method for evaluating the therapeutic effect of a drug for treating adolescent depression according to claim 1, wherein: The step of acquiring physiological index data and behavior data of a depression patient comprises: acquiring a medication start time point of the depression patient, and constructing a medication cycle based on the medication start time point; acquiring heart rate variability and skin electric response data of the patient continuously through a wearable device within the medication cycle, and summarizing the data as physiological index data of the depression patient; recording the social application use frequency, outdoor activity trajectory and sleep monitoring data of the patient through a mobile terminal, scoring the data by combining the Pittsburgh Sleep Quality Index, and obtaining the social activity frequency and sleep quality score; summarizing the social activity frequency and sleep quality score as behavior data of the depression patient, and synchronously storing the behavior data and the physiological index data in a preset data storage dataset. 3.The method for evaluating the therapeutic effect of a drug for treating adolescent depression according to claim 1, characterized in that: The step of normalizing the physiological index data and the behavior data, and performing dimension reduction processing on the normalized data to extract key feature vectors comprises: aligning the physiological index data and the behavior data according to the time sequence, and eliminating the dimensional difference by using the minimum-maximum normalization method; performing covariance analysis on the normalized physiological index data and behavior data, extracting principal component features and reducing the features to a two-dimensional feature space to obtain low-dimensional feature vectors; calculating the Euclidean distance between adjacent time points according to the time continuity of the low-dimensional feature vectors, combining a sliding window mechanism to filter feature points satisfying the distance threshold, and constructing a dynamic change trajectory curve; identifying the curvature mutation point of the trajectory curve as a key turning node, dividing a feature interval centered on the key turning node, and extracting the curvature, slope and change amplitude in the feature interval as key feature vectors. 4.The method for evaluating the therapeutic effect of a drug for treating adolescent depression according to claim 1, characterized in that: The step of acquiring a medication window corresponding to each key feature vector, and quantitatively evaluating the drug efficacy in the medication window to obtain a drug efficacy score comprises: mapping each key feature vector to a time node in the medication cycle, and determining a corresponding medication window interval in the medication cycle, wherein each medication window is extended by a preset duration centered on the time node of the key feature vector. Average the key feature vectors in the medication window and record them as the efficacy evaluation index; Compare the efficacy evaluation index with the efficacy benchmark value, calculate the deviation of the evaluation index from the efficacy benchmark value, and then map the deviation to a preset scoring table to obtain the corresponding drug efficacy score.

5. The method for evaluating the therapeutic effect of a drug for treating adolescent depression according to claim 1, wherein: The step of interpolating and compensating the missing period between adjacent medication windows to form a dynamic efficacy evaluation sequence comprises: Identify the blank time interval between adjacent medication windows and count the length of the blank time interval; Obtain the drug efficacy scores in the two medication windows before and after the blank time interval and input them into a preset efficacy change function, and record the output of the efficacy change function as the interpolation compensation score; Discretely sample the blank time interval in a preset time dimension, assign values to each sampling time point through the interpolation compensation score, and generate an interpolation compensation score sequence; Splice the interpolation compensation score sequence with the score sequence under the medication windows before and after the blank time interval to form a continuous dynamic efficacy evaluation sequence.

6. The method for evaluating the efficacy of a drug for treating adolescent depression according to claim 1, wherein: The step of identifying the efficacy change trend according to the dynamic efficacy evaluation sequence comprises: Time segment the dynamic efficacy evaluation sequence, taking each medication window as the reference unit, and calculate the average rate of change of the efficacy score with time in the medication window; When the average rate continuously increases positively and continuously exceeds N medication windows, it indicates that the drug inhibition is effective, and the efficacy change in the application medication window is marked as an efficacy rising trend; When the average rate continuously changes negatively and continuously exceeds N medication windows, it is determined that the efficacy is weakened or the disease rebounds, and the efficacy change in the application medication window is marked as an efficacy declining trend; When the average rate appears both positive growth and negative change within less than N medication windows, it is determined that the drug efficacy is unstable, and the efficacy change in the application medication window is not marked as an efficacy fluctuation trend.

7. The method for evaluating the therapeutic effect of a drug for treating adolescent depression according to claim 1, wherein: The step of calculating and outputting a comprehensive efficacy stability index according to the duration and alternating frequency of the rising trend, fluctuation trend and declining trend in the medication cycle comprises: Statistically analyze the total duration of the efficacy rising trend, the total duration of the efficacy fluctuation trend and the total duration of the efficacy declining trend in the medication cycle; Calculate the proportion of the total duration of the efficacy rising trend, the total duration of the efficacy fluctuation trend and the total duration of the efficacy declining trend in the medication cycle, and record them as the first proportion coefficient, the second proportion coefficient and the third proportion coefficient respectively; Extract the alternating frequency between the efficacy rising trend, the fluctuation trend and the declining trend, and record it as the trend alternating frequency coefficient; Obtain the time point distribution position of the efficacy rising trend, the efficacy fluctuation trend and the efficacy declining trend in the medication cycle, and match the time sequence weight coefficient of the efficacy rising trend, the efficacy fluctuation trend and the efficacy declining trend in the medication cycle according to the time point distribution position; Weighted sum the first proportion coefficient, the second proportion coefficient, the third proportion coefficient and the trend alternating frequency coefficient, and perform nonlinear mapping combined with the time sequence weight coefficient to generate a comprehensive efficacy stability index. 8.The method for evaluating the therapeutic effect of a drug for treating adolescent depression according to claim 1, characterized in that: The step of comparing the comprehensive efficacy stability index with the preset clinical efficacy standard to determine the efficacy grade of the drug comprises: Obtaining a clinical efficacy classification interval, and comparing the comprehensive efficacy stability index with the clinical efficacy classification interval; If the comprehensive efficacy stability index exceeds the upper limit of the clinical efficacy classification interval, it is determined that the drug efficacy level is significantly improved; If the comprehensive efficacy stability index is within the clinical efficacy classification interval, it is determined that the drug efficacy level is effective; If the comprehensive efficacy stability index is below the lower limit of the clinical efficacy classification interval, it is determined that the drug efficacy level is ineffective, and an early warning prompt is sent to the patient monitoring terminal.

9. A system for evaluating the therapeutic effect of a drug for treating adolescent depression, characterized by: The method for evaluating the efficacy of a drug for treating adolescent depression according to any one of claims 1 to 8, comprising: a data acquisition module for acquiring physiological index data and behavior data of a patient with depression, wherein the physiological index data includes heart rate variability and skin electric response, and the behavior data includes social activity frequency and sleep quality score; a data processing module for normalizing the physiological index data and the behavior data, and performing dimension reduction processing on the normalized data to extract key feature vectors; a quantitative evaluation module for obtaining a drug window corresponding to each key feature vector, and quantitatively evaluating the drug efficacy in the drug window to obtain a drug efficacy score; a trend identification module for arranging the drug windows in a drug time axis according to the occurrence time sequence, interpolating and compensating for the missing period between adjacent drug windows to form a dynamic efficacy evaluation sequence, and identifying the efficacy change trend according to the dynamic efficacy evaluation sequence, wherein the efficacy change trend includes an upward trend, a fluctuation trend and a downward trend; an efficacy evaluation module for calculating and outputting a comprehensive efficacy stability index according to the duration and alternating frequency of the upward trend, the fluctuation trend and the downward trend in the drug cycle, and comparing the comprehensive efficacy stability index with a preset clinical efficacy standard to determine the efficacy level of the drug.

10. An electronic device, comprising: The electronic device comprises: at least one processor; and a memory connected in communication with the at least one processor; wherein the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to execute the method for evaluating the efficacy of a drug for treating adolescent depression according to any one of claims 1 to 8.

Images