Animal model dynamic monitoring method and system
Patent Information
- Application Number
- CN202611080503.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-07-21
- Publication Date
- 2026-08-21
AI Technical Summary
[0004]本发明提供一种动物模型体征动态监测方法及系统，以解决现有的问题
[0015]本发明的技术方案的有益效果是：本发明在同一动物模型在同一检测期内，基于同一维度所含体征监测数据在整体时序上的连续平均数值，分析同一维度在同一检测期内的线性进展变化状态，计算体征阶段推进度；在不割裂时间线的前提下，提取出该维度体征不可逆的核心趋势成分，将经验模糊的感受转化为一个明确可计算的速率指标；然后基于体征阶段推进度与体征监测数据各自在动物模型整体与时序整体之间的相似变化量，分析同一维度与动物模型在同一检测期内时序推进时二者相似关联的强度，计算阶段进度相关系数；使后续综合评估中自动让高相关系数的体征拥有更高参考比重，抑制低相关体征的噪声，将多模态融合从简单平均升级为病理状态引导的加权决策；然后结合阶段进度相关系数以及同一体征隶属区间内体征监测数据的数值分布，分析同一体征隶属区间内体征监测数据的概率密度分布状态，计算不同体征监测数据的体征动态隶属度；将硬性划分升级为概率化的连续描述，从而精细捕捉状态间的微观迁移。本发明通过对整个实验动物模型进行阶段划分后，分析不同阶段下监测数值与所属阶段需要的目标病症的相关性，进而计算体征动态隶属度进行动态监测，将整个疾病模型的多个阶段进行了有效的连续整合，提高了动物模型体征动态监测效率。
Smart Images
Figure CN122604334A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of diagnostic measurement technology, specifically to a method and system for dynamic monitoring of vital signs in animal models. Background Technology
[0002] Heart failure is the end-stage manifestation of cardiovascular disease, and hypertension is one of the main causes of heart failure. Long-term elevated blood pressure can cause cardiac remodeling, eventually leading to total heart failure. Among environmental factors, salt intake is crucial, leading to the concept of "salt sensitivity": high salt intake leads to abnormally high blood pressure. DOCA (deoxycorticosterone acetate)-induced hypertension models belong to this category; without high salt intake, the model does not form. Its basis is related to renal sodium retention and Na+. + K + - This is related to decreased ATPase activity. Given that heart failure remains difficult to cure, an animal model of heart failure accompanied by hypertension was constructed by combining DOCA with a high-salt diet. This model simultaneously exhibits glomerular hypertrophy and renal remodeling, which can provide a stable preclinical basis for the development of new drugs and target screening for heart failure.
[0003] Current monitoring methods primarily rely on researchers using instruments (such as sphygmomanometers, echocardiograms, and blood gas analyzers) to collect various vital signs data (such as blood pressure, heart rate, echocardiographic parameters, respiratory rate, and blood biochemical indicators) from animal subjects. These multiple vital signs data are then used to make subjective judgments of abnormalities. In real-world scenarios, the construction of a disease model is a continuous and multi-stage process. The focus of attention and interpretation of vital signs differs at different pathological stages. Current methods lack an objective and quantitative basis for effectively analyzing the monitoring process of the disease model at different stages, resulting in a significant reduction in the efficiency of dynamic monitoring of animal model vital signs. Summary of the Invention
[0004] This invention provides a method and system for dynamic monitoring of vital signs in animal models to solve existing problems.
[0005] The present invention provides a method and system for dynamic monitoring of vital signs in an animal model, which adopts the following technical solution: This invention proposes a method for dynamic monitoring of vital signs in animal models, which includes the following steps: Acquire several vital sign monitoring data of each animal model under different dimensions within the same testing period; In the same animal model and within the same testing period, based on the continuous average values of the vital signs monitoring data contained in the same dimension over the overall time series, we analyze the linear progression of the same dimension within the same testing period and calculate the progression of the same dimension of vital signs in the same animal model within the same testing period. Based on the similarity of the progress of vital signs and the changes in vital sign monitoring data between the animal model as a whole and the time series, we analyze the strength of the similarity between the two when the same dimension and the animal model progresses in the same detection period, and calculate the correlation coefficient of the progress of the same dimension in the same detection period. Cluster the vital sign monitoring data within the same detection period to obtain several clusters; based on the distance between the cluster centers of adjacent clusters and combined with the correlation coefficient of the stage progress, analyze the uniformity of the distribution density of data content within the same detection period, and calculate several weighted boundary points within the same detection period; divide the interval formed by the vital sign monitoring data within the same detection period into several vital sign membership intervals through the weighted boundary points; Based on the numerical distribution of vital sign monitoring data within the same vital sign membership interval, the probability density distribution of vital sign monitoring data within the same vital sign membership interval is analyzed, and the dynamic membership degree of different vital sign monitoring data is calculated; based on the dynamic membership degree of vital signs, dynamic monitoring and early warning are carried out for each animal model.
[0006] Preferably, the method for obtaining the progress of the vital signs stage is as follows: In the same animal model and during the same testing period, a sequence of vital sign monitoring data of the same dimension over the entire time series is taken as a time series of vital sign monitoring data of the same dimension; by comparing the average difference values of adjacent vital sign monitoring data within the time series of vital sign monitoring data, several continuous linear progression values of the time series of vital sign monitoring data are calculated. The mean of all continuous linear progression values is used as the progression of the same dimension of vital signs in the same animal model during the same testing period.
[0007] Preferably, the method for obtaining the continuous linear progression value is as follows: Obtain the length of the maximum fluctuation interval in the same dimension; for any pair of adjacent vital sign monitoring data, take the ratio of the difference between the pair of vital sign monitoring data to the length of the maximum fluctuation interval as the continuous linear progression value.
[0008] Preferably, the method for obtaining the stage progress correlation coefficient is as follows: A sequence consisting of vital sign monitoring data of all animal models in the same dimension during the same testing period is used as a vital sign monitoring control sequence. The sequence of the progression of vital signs in the same dimension of all animal models within the same detection period was used as the control sequence of the progression of vital signs. The correlation coefficient between the control sequence of the progression of vital signs and the control sequence of vital signs monitoring was calculated and used as the correlation coefficient of the progression of vital signs in the same dimension within the same detection period.
[0009] Preferably, the method for obtaining the vital sign monitoring control sequence is as follows: The mean value of vital sign monitoring data of all animal models at the same time and in the same dimension during the same testing period was used as the comprehensive vital sign monitoring data; the sequence of all comprehensive vital sign monitoring data arranged in time sequence was used as the control sequence for the progression of vital sign stages.
[0010] Preferably, the method for obtaining the weighted boundary point is as follows: By calculating the weighted reference weight between adjacent clusters based on the distance between the cluster centers and the correlation coefficient of the included stage progress, the weighted reference weight is assigned to the cluster centers within the adjacent clusters, and the cluster centers are adjusted to obtain the weighted boundary points within different clusters.
[0011] Preferably, the method for obtaining the weighted reference weights is as follows: The distance between the cluster centers of adjacent clusters is taken as the center distance between adjacent clusters; the mean of the stage progress correlation coefficients contained in each of the two adjacent clusters is taken as the stage progress baseline value of each cluster; the difference between the stage progress baseline values of adjacent clusters is taken as the boundary weight between adjacent clusters; and the product of the center distance and the boundary weight is taken as the weighted reference weight between adjacent clusters.
[0012] Preferably, the method for obtaining the dynamic membership degree of the vital signs is as follows: For any given vital sign membership interval, construct a data histogram of the vital sign membership interval using the vital sign monitoring data within the interval; select the membership function type of the vital sign membership interval by analyzing the data distribution within the histogram; and calculate the dynamic membership degree of different vital sign monitoring data within the vital sign membership interval based on the membership function type.
[0013] Preferably, the method for obtaining the membership function type is as follows: Two membership function types are preset; the product of skewness and standard deviation in the data histogram is used as the reference value for selecting the membership degree of the vital sign membership interval; a threshold condition is determined for the membership selection reference value, and different membership function types are assigned to the vital sign membership interval based on the condition determination result.
[0014] The present invention also proposes an animal model dynamic monitoring system, including a memory and a processor, wherein the processor executes a computer program stored in the memory to implement the steps of the above-described animal model dynamic monitoring method.
[0015] The beneficial effects of the technical solution of this invention are as follows: In the same animal model and within the same detection period, based on the continuous average values of the vital sign monitoring data within the same dimension over the overall time series, this invention analyzes the linear progression of the same dimension within the same detection period and calculates the rate of progression of vital signs. Without disrupting the timeline, it extracts the irreversible core trend components of the vital signs in that dimension, transforming vague, empirical perceptions into a clear and calculable rate indicator. Then, based on the rate of progression of vital signs and the similarity of changes in the vital sign monitoring data between the overall animal model and the overall time series, it analyzes the same dimension and the animal model within the same detection period. This invention calculates the correlation coefficient between two similarities as the time sequence progresses, enabling the automatic allocation of higher-correlation-coefficient signs to subsequent comprehensive assessments and suppressing noise from low-correlation signs. This upgrades multimodal fusion from simple averaging to pathology-guided weighted decision-making. Then, combining the phase progression correlation coefficient with the numerical distribution of sign monitoring data within the same sign's membership interval, the probability density distribution of sign monitoring data within the same sign's membership interval is analyzed, and the dynamic membership degree of different sign monitoring data is calculated. This upgrades rigid division to probabilistic continuous description, thereby precisely capturing microscopic shifts between states. By dividing the entire experimental animal model into stages, analyzing the correlation between monitoring values at different stages and the target disease required for that stage, and then calculating the dynamic membership degree of signs for dynamic monitoring, this invention effectively and continuously integrates multiple stages of the entire disease model, improving the efficiency of dynamic sign monitoring in animal models. Attached Figure Description
[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0017] Figure 1 This is a flowchart illustrating the steps of a method for dynamic monitoring of vital signs in an animal model according to the present invention. Detailed Implementation
[0018] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a method and system for dynamic monitoring of vital signs in an animal model according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.
[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.
[0020] The following description, in conjunction with the accompanying drawings, details the specific scheme of the method and system for dynamic monitoring of vital signs in an animal model provided by the present invention.
[0021] Please see Figure 1 The diagram illustrates a flowchart of a method for dynamic monitoring of vital signs in an animal model according to an embodiment of the present invention. The method includes the following steps: Step S001: Obtain several vital sign monitoring data of each animal model under different dimensions within the same detection period.
[0022] It should be noted that current monitoring methods primarily rely on researchers using instruments (such as sphygmomanometers, echocardiograms, and blood gas analyzers) to collect various vital signs data (such as blood pressure, heart rate, echocardiogram parameters, respiratory rate, and blood biochemical indicators) from animal subjects, and then making human judgments based on a combination of multiple vital sign data. In real-world scenarios, the construction of a disease model is a continuous and multi-stage process. The focus of vital sign information that needs to be monitored and interpreted differs at different pathological stages. Current methods lack an objective and quantitative basis for effectively analyzing the monitoring process of the disease model at different stages, resulting in a significant reduction in the efficiency of dynamic monitoring of animal model vital signs.
[0023] In one specific implementation of this invention, the process of acquiring vital sign monitoring data is as follows: extracting data from a vital sign database of the same type of animal model. An animal model in Within each testing period Several vital sign monitoring data under various dimensions.
[0024] It should be noted that this embodiment uses a mouse as an animal model for illustration, and the vital signs monitoring data is recorded at a frequency of 1 time per second. indivual, indivual, This will be described using an example. Furthermore, this embodiment uses the induction period, compensatory period, transition period, decompensation period, and terminal period as examples. This example uses a single testing period as an example; this embodiment uses systolic blood pressure, diastolic blood pressure, heart rate, left ventricular mass index, left ventricular end-diastolic volume, ejection fraction, serum brain natriuretic peptide concentration, respiratory rate, and activity level as indicators. This example uses one dimension as an illustration; however, it is not specifically limited to any particular dimension. , , It depends on the specific implementation situation.
[0025] It should be noted that the induction period refers to the stage from the start of surgery / drug administration until a sustained and significant increase in blood pressure occurs. The monitoring focus is on the initial upward trend of blood pressure and renal-related indicators. The compensatory period refers to the stage where blood pressure continues to rise, and the heart begins to develop concentric hypertrophy to maintain pumping function, but without obvious heart failure. The monitoring focus is on the left ventricular mass index, stable high blood pressure, and early changes in diastolic function parameters. The transition period refers to the stage where the cardiac compensation mechanism is no longer sufficient to fully compensate, diastolic function deteriorates significantly, and there may be a slight decrease in activity or an increase in water intake. The monitoring focus is on rapid changes in diastolic function parameters and slight changes in behavioral indicators. The decompensated period refers to the stage where respiratory rate increases sharply. The monitoring focus is on respiratory rate and the presence of direct or indirect evidence of pulmonary edema / ascites. The end-stage refers to a severe state of heart failure, which may be accompanied by cachexia and multiple organ failure. The monitoring focus is on systemic indicators and survival status. The above five detection periods are based on the accepted pathological process of the DOCA+ salt-induced HFpEF model and are obtained through empirical division. The animal model vital signs database will automatically divide the detection periods using the built-in intelligent classification model when recording vital signs monitoring data, which will not be described in this embodiment.
[0026] It should be noted that the vital sign monitoring data in this embodiment have all been normalized by default. This embodiment assumes that the data is normalized by default. Normalization is performed using a function as an example.
[0027] Thus, the above methods have yielded several vital sign monitoring data for each animal model under different dimensions during the same testing period.
[0028] Step S002: In the same animal model and within the same testing period, based on the continuous average value of the vital signs monitoring data contained in the same dimension over the overall time series, analyze the linear progression of the same dimension in the same testing period, and calculate the progression of the vital signs in the same dimension in the same animal model within the same testing period.
[0029] It's important to note that single-point collected vital sign data contains numerous reversible fluctuations due to circadian rhythms, food intake, and transient stress. Directly using raw values would obscure the true pathological deterioration signals. Employing continuous averaging over the entire time series (such as moving averages or cumulative averages) allows for the extraction of the irreversible core trend components of that dimension of the vital sign without disrupting the timeline, providing a clean "deterioration baseline" for subsequent analysis. Analyzing the linear progression throughout the entire monitoring period essentially involves calculating the directional slope of this trend line. For DOCA-induced hypertension and heart failure, decreased heart rate variability, reduced nocturnal activity, and weakened body temperature rhythm often manifest as a slow but steady directional drift. The steepness of this slope directly quantifies the rate of organ function loss, transforming the vague feeling of "the disease is worsening" into a clear and calculable rate indicator.
[0030] Preferably, in some implementations of the present invention, the method for obtaining the progression of vital signs in the same stage is as follows: Within the same animal model and the same testing period, a sequence of vital sign monitoring data of the same dimension over the entire time series is taken as a time series sequence of vital sign monitoring data of the same dimension; the average difference values of adjacent vital sign monitoring data within the time series sequence are compared to calculate several continuous linear progression values of the time series sequence; the mean of all continuous linear progression values is taken as the progression of the vital sign in the same dimension within the same animal model and the same testing period. The specific process is as follows: In the same animal model and within the same testing period, the sequence of all vital sign monitoring data of the same dimension arranged in chronological order from early to late is taken as the time series sequence of vital sign monitoring data of the same dimension.
[0031] Preferably, in some implementations of the present invention, the method for obtaining the continuous linear progression value is as follows: obtaining the length of the maximum fluctuation interval in the same dimension; for any pair of adjacent vital sign monitoring data, the ratio of the difference between the pair of vital sign monitoring data to the length of the maximum fluctuation interval is taken as the continuous linear progression value. The specific process is as follows: As an example, the value of continuous linear progression can be calculated using the following formula. :
[0032] In the formula, Indicates the first Individual trait monitoring data and the first Continuous linear progression values between individual trait monitoring data; Indicates the first Individual trait monitoring data and the first The absolute value of the difference between individual trait monitoring data; This represents the maximum preset monitoring value for the same dimension. In this embodiment, the preset value is... Let's take an example to illustrate; This represents the minimum preset monitoring value for the same dimension. In this embodiment, the preset value is... Let's take an example to illustrate; Indicates the first Individual trait monitoring data; Indicates the first Individual trait monitoring data; Indicates taking the absolute value; This indicates the length of the maximum fluctuation range in the same dimension.
[0033] Furthermore, all continuous linear progression values of the time series of the vital sign monitoring data are obtained, and the mean of all continuous linear progression values is used as the progression of the vital sign stage in the same dimension within the same detection period of the same animal model.
[0034] Thus, the above method was used to obtain the progression of the same dimension of vital signs in the same animal model during the same testing period.
[0035] Step S003: Based on the similarity changes of the progress of the physical signs stage and the monitoring data of the physical signs between the animal model as a whole and the time series, analyze the strength of the similarity correlation between the two when the same dimension and the animal model progresses in the same detection period, and calculate the correlation coefficient of the stage progress of the same dimension in the same detection period.
[0036] It's important to note that the progression of disease is a pre-defined trajectory of deterioration, while the raw vital signs data contain various fluctuations. The correlation coefficient between the two directly reflects the extent to which each shift in the vital sign is synchronized with the disease progression. A higher coefficient indicates that the vital sign can serve as a highly reliable surrogate indicator of disease progression; conversely, a lower coefficient suggests that it is too susceptible to interference from non-disease factors and is not suitable as an independent criterion. Based on this, the system can achieve adaptive dynamic weighting, automatically giving higher weight to vital signs with high correlation coefficients in comprehensive assessments and suppressing noise from low-correlation vital signs, upgrading multimodal fusion from simple averaging to pathology-guided weighted decision-making. Furthermore, such dramatic changes in correlation coefficients are themselves sensitive signals of disease progression. For example, when the correlation coefficient between blood pressure and progression drops sharply, while the correlation coefficient between respiration or activity levels spikes, it signifies the collapse of compensatory mechanisms and the onset of decompensation, which can drive a seamless shift in monitoring focus.
[0037] Preferably, in some implementations of the present invention, the method for obtaining the correlation coefficient of stage progress is as follows: a sequence composed of vital sign monitoring data of all animal models in the same dimension within the same testing period is used as a vital sign monitoring control sequence; a sequence composed of the stage progress of all animal models in the same dimension within the same testing period is used as a vital sign stage progress control sequence; the correlation coefficient between the vital sign stage progress control sequence and the vital sign monitoring control sequence is calculated and used as the stage progress correlation coefficient of the same dimension within the same testing period. The specific process is as follows: Preferably, in some implementations of the present invention, the method for obtaining the vital sign monitoring control sequence is as follows: the mean value of vital sign monitoring data of all animal models at the same time and in the same dimension during the same detection period is used as the comprehensive vital sign monitoring data; the sequence formed by arranging all comprehensive vital sign monitoring data in time sequence is used as the vital sign stage progress control sequence. The specific process is as follows: The mean of the vital sign monitoring data of all animal models at the same time and in the same dimension within the same testing period is used as the comprehensive vital sign monitoring data at the same time and in the same dimension within the same testing period. Comprehensive vital sign monitoring data at all times within the same dimension within the same testing period are obtained, and these comprehensive vital sign monitoring data are arranged in chronological order from earliest to latest as a control sequence for the progression of vital signs within the same dimension within the same testing period.
[0038] Furthermore, the sequence of the same dimension of physical sign progression in all animal models within the same detection period is used as the physical sign progression control sequence; the Pearson correlation coefficient between the physical sign progression control sequence and the physical sign monitoring control sequence is used as the stage progress correlation coefficient within the same dimension of the same detection period.
[0039] It should be noted that if the lengths of the control sequence for the progression of vital signs and the control sequence for monitoring vital signs are inconsistent, the shorter sequence will be padded with zeros until the lengths of the two sequences are consistent. In addition, the process of obtaining the Pearson correlation coefficient is well known and will not be described in this embodiment.
[0040] Thus, the correlation coefficient of the same dimension of stage progress within the same detection period was obtained using the above method.
[0041] Step S004: Cluster the vital sign monitoring data within the same detection period to obtain several clusters; based on the distance between the cluster centers of adjacent clusters and combined with the correlation coefficient of the stage progress, analyze the uniformity of the distribution density of data content within the same detection period, and calculate several weighted boundary points within the same detection period; divide the interval formed by the vital sign monitoring data within the same detection period into several vital sign membership intervals through the weighted boundary points.
[0042] It should be noted that clustering can discover naturally occurring steady-state patterns in data without pre-defined constraints, with each cluster representing a typical combination of physical signs. Introducing a stage-progression correlation coefficient to weight the distance between adjacent cluster centers allows the determination of state boundaries to focus more on feature changes strongly correlated with disease progression, respecting the data structure while strengthening pathological indications. The weighted boundary points calculated accordingly divide the data space of the entire detection period into several physical sign membership intervals, each interval corresponding to a clearly defined disease functional state.
[0043] The absolute value of the difference between different vital sign monitoring data is used as the distance metric. Based on the distance metric, all vital sign monitoring data of all animal models within the same detection period are clustered to obtain several clusters.
[0044] It should be noted that this embodiment uses the K-means clustering algorithm as an example to obtain clusters, and the number of categories in the K-means clustering algorithm is specified. This embodiment is pre-set Taking distance as an example, based on distance measurement and The process of obtaining clusters is a well-known part of the K-means clustering algorithm, and will not be described in detail in this embodiment.
[0045] Preferably, in some implementations of the present invention, the method for obtaining the weighted boundary point is as follows: a weighted reference weight is calculated between adjacent clusters based on the distance between the cluster centers and the correlation coefficient of the included stage progress; the weighted reference weight is assigned to the cluster centers within the adjacent clusters, and the cluster centers are adjusted to obtain the weighted boundary points within different clusters. The specific process is as follows: Preferably, in some implementations of the present invention, the method for obtaining the weighted reference weight is as follows: the distance between the cluster centers of adjacent clusters is taken as the center distance between adjacent clusters; in two adjacent clusters, the mean of the stage progress correlation coefficients contained in each cluster is taken as the stage progress baseline value of each cluster; the difference between the stage progress baseline values between adjacent clusters is taken as the boundary weight between adjacent clusters; and the product of the center distance and the boundary weight is taken as the weighted reference weight between adjacent clusters. The specific process is as follows: The mean of all vital sign monitoring data within each cluster is used as the cluster center value for each cluster; the mean of all stage progress correlation coefficients within each cluster is used as the stage progress baseline value for each cluster.
[0046] Furthermore, as an example, the weighted reference weight can be calculated using the following formula. :
[0047] In the formula, This represents the weighted reference weight between adjacent clusters; This represents the absolute value of the difference between the cluster center values of adjacent clusters, i.e., the center distance; The absolute value of the difference between the stage progress baseline values between adjacent clusters is called the boundary weight. This represents the normalization function.
[0048] Furthermore, taking any cluster as an example, the mean of the weighted reference weights between this cluster and its two adjacent clusters is used as the boundary weight of this cluster; the product of the cluster center value and the boundary weight is used as the weighted boundary point within this cluster. The weighted boundary points within each cluster are obtained; each weighted boundary point is used as a dividing line to divide the largest numerical interval of vital sign detection data from all animal models within the same detection period, and each of these intervals is used as the vital sign membership interval for the same detection period. The membership interval for each vital sign within the same detection period is then obtained.
[0049] Thus, the interval to which each vital sign belongs in the same detection period is obtained through the above method.
[0050] Step S005: Based on the numerical distribution of vital sign monitoring data within the same vital sign membership interval, analyze the probability density distribution of vital sign monitoring data within the same vital sign membership interval, calculate the dynamic membership degree of different vital sign monitoring data; and perform dynamic monitoring and early warning for each animal model based on the dynamic membership degree of vital signs.
[0051] It's important to note that after defining the membership intervals of vital signs, calculating the dynamic membership degree transforms the rigid "belonging / not belonging" classification into a probabilistic, continuous description of "to what extent one belongs," thus precisely capturing the micro-transitions between states. By analyzing the probability density distribution of vital sign data within an interval, the "core area" and "edge zone" of that state can be identified. This allows each monitoring session to obtain membership degrees for multiple states (e.g., 0.8 for compensated state, 0.2 for mild decompensated state), truly reflecting the ambiguous "both this and that" attributes of the disease transition period. The gradual changes and intersections of the dynamic membership degree curve can reveal early signs of state switching earlier and more smoothly than fixed thresholds or hard labels, forming a sensitive early warning signal for disease progression.
[0052] Preferably, in some implementations of the present invention, the method for obtaining the dynamic membership degree of vital signs is as follows: for any vital sign membership interval, a data histogram of the vital sign membership interval is constructed using the vital sign monitoring data within the interval; by analyzing the data distribution state within the histogram, the membership function type of the vital sign membership interval is selected; and the dynamic membership degree of different vital sign monitoring data within the interval is calculated using the membership function type. The specific process is as follows: Taking any vital sign interval within the same detection period as an example, the magnitude of the vital sign monitoring data is used as the horizontal axis, and the number of corresponding vital sign monitoring data is used as the vertical axis. Through all the vital sign monitoring data within the interval, the horizontal axis and the vertical axis, a numerical histogram of the vital sign interval is constructed.
[0053] Preferably, in some implementations of the present invention, the method for obtaining the membership function type is as follows: two membership function types are preset; the product of the skewness and standard deviation within the data histogram is used as the membership selection reference value for the vital sign membership interval; a threshold condition is determined on the membership selection reference value, and different membership function types are assigned to the vital sign membership interval based on the condition determination result. The specific process is as follows: The normalized value of the product of skewness and standard deviation within the data histogram is used as the reference value for selecting the membership degree of the trait's membership interval; if the membership degree selection reference value is greater than a preset threshold... The trigonometric function is used as the membership function type for the membership interval of this feature; if the membership selection reference value is less than or equal to the preset threshold... The trapezoidal function is used as the membership function type for the membership interval of this feature. In this embodiment, the trapezoidal function is used as... This example is used for illustration; no specific limitations are set in this embodiment. It depends on the specific implementation situation.
[0054] It should be noted that the triangular function and the trapezoidal function are two membership function types preset in this embodiment, and both of these functions are existing well-known function models, which will not be described in detail in this embodiment.
[0055] It should be noted that if the membership degree reference value is greater than the preset threshold... This indicates that the data curves within the histogram tend to have a "peaked" / "triangular" distribution. Therefore, a triangular function is chosen as the membership function type for this characteristic's membership interval. If the membership selection reference value is less than or equal to the preset threshold... This indicates that the more the data curves formed within the data histogram tend to be "horizontal" or "flat", the more appropriate the trapezoidal function should be as the membership function type for the interval to which this characteristic belongs.
[0056] Furthermore, for each vital sign monitoring data point within the vital sign membership interval, the membership function type of that interval is input, and the output value is used as the dynamic membership degree of each vital sign monitoring data point within that interval. The dynamic membership degrees of the same animal model within the same testing period are input into the built-in monitoring report module of this embodiment to generate a dynamic monitoring report for the same animal model within the same testing period. The process of generating the dynamic monitoring report through the intelligent module is a well-known existing technique and will not be described further in this embodiment.
[0057] It should be noted that the dynamic monitoring report example in this embodiment is roughly as follows: Animal model ID: Rat_2024_05; Monitoring period: 42 days post-surgery; Stage description: Induction period 0.8, Compensation period 0.9, Transition period 0.7, Decompensation period 0, Terminal period 0; Details of vital signs and condition: Systolic blood pressure (SBP): 198 mmHg -> Status: High (Membership: 0.92); Respiratory rate (RF): 125 bpm -> Status: Normal to high (membership: 0.65), Normal (membership: 0.35); Left ventricular mass (LVM): 1.15g -> Status: Medium (Membership: 0.88); Overall assessment: Diastolic function has deteriorated significantly, indicating a late transition period; close monitoring of respiratory rate changes is necessary.
[0058] The above steps complete the method for dynamic monitoring of vital signs in animal models.
[0059] Another embodiment of the present invention provides an animal model vital signs dynamic monitoring system, the system including a memory and a processor, wherein when the processor executes the computer program stored in the memory, it performs the above method steps S001 to S005.
[0060] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.
Claims
1. A method for dynamic monitoring of vital signs in an animal model, characterized in that, The method includes the following steps: Acquire several vital sign monitoring data of each animal model under different dimensions within the same testing period; In the same animal model and within the same testing period, based on the continuous average values of the vital signs monitoring data contained in the same dimension over the overall time series, we analyze the linear progression of the same dimension within the same testing period and calculate the progression of the same dimension of vital signs in the same animal model within the same testing period. Based on the similarity of the progress of vital signs and the changes in vital sign monitoring data between the animal model as a whole and the time series, we analyze the strength of the similarity between the two when the same dimension and the animal model progresses in the same detection period, and calculate the correlation coefficient of the progress of the same dimension in the same detection period. Cluster the vital sign monitoring data within the same detection period to obtain several clusters; based on the distance between the cluster centers of adjacent clusters and combined with the correlation coefficient of the stage progress, analyze the uniformity of the distribution density of data content within the same detection period, and calculate several weighted boundary points within the same detection period; divide the interval formed by the vital sign monitoring data within the same detection period into several vital sign membership intervals through the weighted boundary points; Based on the numerical distribution of vital sign monitoring data within the same vital sign membership interval, the probability density distribution of vital sign monitoring data within the same vital sign membership interval is analyzed, and the dynamic membership degree of different vital sign monitoring data is calculated; based on the dynamic membership degree of vital signs, dynamic monitoring and early warning are carried out for each animal model.
2. The method for dynamic monitoring of vital signs in an animal model according to claim 1, characterized in that, The method for obtaining the progress of the vital signs stage is as follows: In the same animal model and during the same testing period, a sequence of vital sign monitoring data of the same dimension over the entire time series is taken as a time series of vital sign monitoring data of the same dimension; by comparing the average difference values of adjacent vital sign monitoring data within the time series of vital sign monitoring data, several continuous linear progression values of the time series of vital sign monitoring data are calculated. The mean of all continuous linear progression values is used as the progression of the same dimension of vital signs in the same animal model during the same testing period.
3. The method for dynamic monitoring of vital signs in an animal model according to claim 2, characterized in that, The method for obtaining the continuous linear progression value is as follows: Obtain the length of the maximum fluctuation interval in the same dimension; for any pair of adjacent vital sign monitoring data, take the ratio of the difference between the pair of vital sign monitoring data to the length of the maximum fluctuation interval as the continuous linear progression value.
4. The method for dynamic monitoring of vital signs in an animal model according to claim 1, characterized in that, The method for obtaining the correlation coefficient of the stage progress is as follows: A sequence consisting of vital sign monitoring data of all animal models in the same dimension during the same testing period is used as a vital sign monitoring control sequence. The sequence of the progression of vital signs in the same dimension of all animal models within the same detection period was used as the control sequence of the progression of vital signs. The correlation coefficient between the control sequence of the progression of vital signs and the control sequence of vital signs monitoring was calculated and used as the correlation coefficient of the progression of vital signs in the same dimension within the same detection period.
5. The method for dynamic monitoring of vital signs in an animal model according to claim 4, characterized in that, The method for obtaining the vital signs monitoring control sequence is as follows: The mean value of vital sign monitoring data of all animal models at the same time and in the same dimension during the same testing period was used as the comprehensive vital sign monitoring data; the sequence of all comprehensive vital sign monitoring data arranged in time sequence was used as the control sequence for the progression of vital sign stages.
6. The method for dynamic monitoring of vital signs in an animal model according to claim 1, characterized in that, The method for obtaining the weighted boundary point is as follows: By calculating the weighted reference weight between adjacent clusters based on the distance between the cluster centers and the correlation coefficient of the included stage progress, the weighted reference weight is assigned to the cluster centers within the adjacent clusters, and the cluster centers are adjusted to obtain the weighted boundary points within different clusters.
7. The method for dynamic monitoring of vital signs in an animal model according to claim 6, characterized in that, The method for obtaining the weighted reference weights is as follows: The distance between the cluster centers of adjacent clusters is taken as the center distance between adjacent clusters; the mean of the stage progress correlation coefficients contained in each of the two adjacent clusters is taken as the stage progress baseline value of each cluster; the difference between the stage progress baseline values of adjacent clusters is taken as the boundary weight between adjacent clusters; and the product of the center distance and the boundary weight is taken as the weighted reference weight between adjacent clusters.
8. The method for dynamic monitoring of vital signs in an animal model according to claim 1, characterized in that, The method for obtaining the dynamic membership degree of the vital signs is as follows: For any given vital sign membership interval, construct a data histogram of the vital sign membership interval using the vital sign monitoring data within the interval; select the membership function type of the vital sign membership interval by analyzing the data distribution within the histogram; and calculate the dynamic membership degree of different vital sign monitoring data within the vital sign membership interval based on the membership function type.
9. The method for dynamic monitoring of vital signs in an animal model according to claim 8, characterized in that, The method for obtaining the membership function type is as follows: Two membership function types are preset; the product of skewness and standard deviation in the data histogram is used as the reference value for selecting the membership degree of the vital sign membership interval; a threshold condition is determined for the membership selection reference value, and different membership function types are assigned to the vital sign membership interval based on the condition determination result.
10. A dynamic monitoring system for vital signs of an animal model, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the computer program is executed by the processor, it implements the steps of the method for dynamic monitoring of vital signs in an animal model as described in any one of claims 1-9.