Short figure development multi-dimensional data fusion evaluation system based on enteric-cerebral-hepatic axis quantitative model

Abstract

The invention relates to the field of information systems and cloud platforms for interaction detection of proteins and metabolites and health correlation in biological material test and analysis, and discloses a short figure development multi-dimensional data fusion evaluation system based on an intestinal brain liver axis quantitative model. Comprising a data acquisition module, an axial gating mapping module and a growth steady state evaluation module, and the data acquisition module acquires an intestinal metabolism component, a liver growth axis component and a neuroendocrine component of a subject; the axial gating mapping module is used for determining signal transduction time delay, calling an intestinal metabolism component at the T-tau moment as a parameter reference through asynchronous offset processing, and determining a gating calibration coefficient in combination with a neuroendocrine component fluctuation slope; and the growth steady state evaluation module performs gain compensation on the liver growth axis component according to the calibration coefficient and outputs a growth and development evaluation index. According to the invention, asynchronous interference of biochemical signal circulation is solved, evaluation model drift caused by phase dislocation is eliminated, and the accuracy of development blocking reason screening is enhanced.

Description

Technical Field

[0001] This invention relates to a multidimensional data fusion evaluation system for short stature development based on a quantitative model of the gut-brain-liver axis, belonging to the technical field of information systems and cloud platforms for detecting protein-metabolite interactions and health-related aspects in biomaterial testing and analysis. Background Technology

[0002] Current developmental assessment systems employ a technical approach that combines growth curve fitting with comparison of endocrine components. These systems are based on the assumption that each biochemical component is independent. They determine the developmental status of subjects by constructing a static assessment model on a cloud server. However, in the testing of complex biological samples, it is difficult to reflect the dynamic interaction characteristics of protein components under complex physiological axes and eliminate the time delay interference of intestinal metabolites on the expression of liver-derived growth proteins. This leads to deviations in the determination of causal chains by traditional biological testing methods.

[0003] However, the physiological feedback process of the gut-brain-liver axis has a physical transmission delay, resulting in a phase difference in the regulatory signals of intestinal metabolites on the expression of liver-derived growth proteins. There is a phase deviation between intestinal metabolic characteristics and liver growth axis responses caused by biological cycles. Because traditional systems select heterogeneous data at the same sampling time and ignore the asynchronous nature of physiological signal flow, the evaluation model will have a logical bias. To address the above bottlenecks, existing improvement ideas focus on improving data processing accuracy or increasing feature dimensions. Specifically, existing technologies have the following shortcomings: 1. The evaluation logic lacks causal alignment in the time dimension. The system selects organ index components at the same time stamp, which cannot reflect the transmission time of biochemical substances in the enterohepatic circulation; 2. The system lacks a quantitative model of axial signal gating mapping, making it difficult to identify the feedback inhibition of source organ signal transmission to target organ, resulting in evaluation errors. Simply increasing the computational scale does not eliminate the asynchronous interference of heterogeneous signals at the principle level. On the contrary, the increased data coupling leads to ambiguity in the determination of causal chains.

[0004] Therefore, how to construct a quantitative model to compensate for the time delay of biochemical signal transduction and establish the causal alignment of asynchronous features, thereby achieving an accurate evaluation of the biochemical association between growth hormone axis-related proteins and intestinal metabolites, has become the technical problem to be solved by this invention. Summary of the Invention

[0005] To address the problems mentioned in the background art, the technical solution of this invention is as follows: A multi-dimensional data fusion evaluation system for short stature development based on a gut-brain-liver axis quantification model, the system comprising a data acquisition module, an axial gating mapping module, and a growth homeostasis evaluation module: The data acquisition module is used to acquire the subject's intestinal metabolic components, liver growth axis components, and neuroendocrine components; The axial gating mapping module, connected to the data acquisition module, is used to determine the transmission delay τ of intestinal metabolic signals to the liver and brain based on the subject's physiological characteristics. The axial gating mapping module also performs asynchronous offset processing, retrieving the intestinal metabolic component at time T-τ as the parameter baseline for the current time T, and calculating the deviation ΔG of the intestinal metabolic component at time T-τ relative to the subject's historical mean. Furthermore, the axial gating mapping module extracts the oscillation slope of the neuroendocrine component at time T. And based on the deflection ΔG and the oscillation slope And the gate calibration coefficient is calculated using the preset response weighting coefficients Φ and θ. ; The growth steady-state evaluation module, connected to the axial gating mapping module, is used to perform gain compensation on the liver growth axis components at time T according to the gating calibration coefficient γ, and to evaluate the liver growth axis components after gain compensation. Data is fused with neuroendocrine components to generate and output evaluation indicators characterizing the homeostasis of the subject's growth and development.

[0006] Preferably, the axial gating mapping module determines the conduction delay. The following steps are followed: Step S1, obtain the subject's age calibration factor σ; Step S2, retrieve the biological index relationship model and establish the mapping relationship between the age calibration factor σ and the baseline time delay constant; Step S3, perform correction on the baseline time delay constant based on the subject's height deviation to generate the conduction time delay τ.

[0007] Preferably, when the axial gating mapping module extracts the deflection ΔG, it is used to calculate the Euclidean distance between the current short-chain fatty acid content and the mean level of subjects at the same developmental stage in the database, and convert the Euclidean distance into the deflection ΔG; wherein, the intestinal metabolic component includes the short-chain fatty acid content and the intestinal microecological characteristic value sequence.

[0008] Preferably, the system further includes a feedback calibration module; the feedback calibration module is connected to the growth homeostasis evaluation module, and is used to obtain the bile acid metabolism characteristics of the subject as the trigger source and use them as the internal arbitration signal, and to perform confidence verification on the evaluation indicators according to the fluctuation phase of the bile acid metabolism characteristics.

[0009] Preferably, the system also includes a dynamic response stress test module; the dynamic response stress test module is used to obtain the growth slope of the intestinal metabolic component and the response delay of the liver growth axis component after the subject inputs a standardized nutritional load, and to correct the transmission weight parameters in the growth homeostasis evaluation module accordingly.

[0010] Preferably, when the growth homeostasis evaluation module performs gain compensation, it is used to adjust the response threshold of the liver growth factor receptor according to the gating calibration coefficient γ, so as to establish a nonlinear regulatory constraint on the sensitivity of intestinal metabolites to the growth axis.

[0011] Preferably, the data acquisition module is also connected to the cloud data synchronization module; the cloud data synchronization module is used to retrieve the subject's historical evaluation trajectory and align the execution sequence of each physiological component with the intestinal stimulation characteristics in the historical spatiotemporal context.

[0012] Preferably, the neuroendocrine component includes a growth hormone-releasing hormone component and a neuronal activity inhibitory factor component; the axial gating mapping module calculates the fluctuation slope. At time T, the first derivative of the growth hormone-releasing hormone component at time T is extracted, and the signal base noise generated by the neuronal activity inhibitory factor component is filtered out.

[0013] Preferably, the growth steady-state evaluation module uses the following formula to calculate the evaluation index. : ,in, For the generated evaluation indicators, The liver growth axis components after gain compensation. The neuroendocrine component is represented by γ, which is the gating calibration coefficient. , and This is a preset weighting factor.

[0014] Preferably, the system also includes a display terminal; the display terminal is connected to the growth steady-state evaluation module and is used to generate the growth trajectory curve of the subject within a preset period according to the evaluation indicators, and to identify the phase deviation of the growth trajectory curve relative to the standard growth curve.

[0015] Compared with the prior art, the beneficial effects of the present invention are: 1. In the quantitative model of the gut-brain-liver axis, an axial gating mapping mechanism for heterogeneous biochemical signals is established to eliminate the isolation bias between various evaluation dimensions. The system extracts the activity component from intestinal metabolic data through the axial gating mapping module and generates gating calibration coefficients for neuroendocrine signals and liver growth axis signals. Dynamic gain adjustment is performed on the signal output of target organs through intestinal metabolic state. At the logical level, the constraint effect of the intestinal environment on the liver growth factor synthesis response is simulated, thereby accurately quantifying the intervention effect of specific metabolites on the expression of growth axis proteins. This enables in-depth analysis of the protein interaction logic in biological samples, thereby improving the accuracy of threshold determination for specific biochemical markers in biomaterial testing and analysis. Compared with the linear evaluation method of traditional systems that only perform threshold comparison on a single endocrine indicator, this invention uses this non-linear logic valve to effectively avoid evaluation omissions caused by ignoring the interaction between organs, ensuring that the evaluation report can reflect the true systemic homeostasis of growth and development.

[0016] 2. An asynchronous fusion correction logic based on physiological conduction delay is introduced to establish causal phase alignment of multi-source data. The system determines the logical conduction delay of intestinal metabolic signals to the liver and brain based on the individual physiological parameters of the subjects, and performs time-series asynchronous offset processing on the collected intestinal metabolic components. This processing simulates the actual transmission time of biochemical substances in enterohepatic circulation and blood circulation on the logical time axis, so that the liver and brain response data at the current moment can be accurately aligned with the intestinal excitation characteristics in the corresponding spatiotemporal background. This solves the long-standing problem of asynchronous interference in biochemical signal flow in health information systems, eliminates model drift caused by the same data sampling time but misaligned biological logical phase, and improves the technical certainty for screening the causes of developmental delay.

[0017] 3. A standardized dynamic response stress test model under load stimulation is constructed to identify latent conduction disorders. By acquiring multi-time-point sequence data after standardized nutrient input, the system extracts the growth slope of intestinal microecological characteristics and the response delay of liver growth factors, and calculates the axial conduction rate component accordingly. This expands the evaluation dimension from static reading analysis to dynamic functional stress testing, enabling the identification of pseudo-normal indicators exhibited by subjects in the resting state due to the body's inherent compensatory mechanisms. By quantifying the axial conduction quality during energy flow fluctuations, the system can accurately identify the dynamic bottlenecks of the growth axis at the execution level. This is used to assess the real-time response characteristics of protein synthesis efficiency under complex biochemical environments, providing clinically more effective early warning of developmental impairment, and effectively revealing deep growth factor receptor bottlenecks such as insufficient growth axis reserve function. This achieves in-depth investigation from the molecular metabolism level to growth and development. Attached Figure Description

[0018] Figure 1 This is a logic diagram of multi-source data fusion for the gut-brain-liver axis with spatiotemporal phase alignment, as described in this invention. Figure 2 This is a breakdown diagram of the functional branches and technical objectives of the short stature development assessment system of the present invention. Detailed Implementation

[0019] The present invention will be further described below with reference to the accompanying drawings and specific embodiments, so that those skilled in the art can understand and implement the present invention. The embodiments are not intended to limit the present invention. In the absence of conflict, the embodiments of the present invention and the features in the embodiments can be combined with each other. The following embodiments are intended for explanation and illustration, and are not intended to limit the scope of protection of the present invention.

[0020] This invention provides a multidimensional data fusion evaluation system for short stature development based on a gut-brain-liver axis quantitative model. It integrates a data acquisition module, an axial gating mapping module, and a growth homeostasis evaluation module. The data acquisition module acquires the intestinal metabolic components, liver growth axis components, and neuroendocrine components of the subject. The axial gating mapping module determines the signal transmission delay τ. This parameter is determined by retrieving a discrete biological time delay mapping table built into the system. The system acquires intestinal metabolic data at a constant frequency of 50Hz through the data acquisition module and stores it in a circular FIFO buffer with a depth of 810,000 data points. During retrieval calibration, the subject's age calibration factor is used as the input index to retrieve the baseline transmission time for a population with a step resolution of 0.5 years from memory. The system reads the current height deviation value. If the height deviation deviates from the standard value by 0.1 standard deviations, the system automatically adds or subtracts 900 seconds from the baseline transmission time to obtain a specific transmission delay in seconds. The intestinal metabolic components at time T-τ are retrieved as the parameter benchmark through asynchronous offset processing and combined with the neuroendocrine components... The gating calibration coefficient γ is determined by the slope of the secretory component fluctuation. The growth homeostasis evaluation module performs gain compensation on the liver growth axis component based on the gating calibration coefficient γ and outputs growth and development evaluation indicators. This system solves the asynchronous interference of biochemical signal flow and eliminates evaluation model drift caused by phase misalignment. To address the application obstacle of the independent and uncorrelated biochemical indicators in the short stature development evaluation process, the data acquisition module connects to the in vitro sample detection device through an application programming interface to obtain the subject's multi-dimensional biochemical parameters. The intestinal metabolic components collected by the data acquisition module include the short-chain fatty acid content measured by chromatography-mass spectrometry and the intestinal microecological characteristic value sequence obtained by sequencing equipment. The neuroendocrine components include the growth hormone-releasing hormone component and the neuronal activity inhibitory factor component. The liver growth axis component includes the serum insulin-like growth factor-I concentration. After receiving the raw data stream, the data acquisition module performs de-identification processing and uses a hash algorithm to perform one-way encryption on the subject's identity information, retaining only the in vitro biochemical values ​​as input for subsequent logical operations, ensuring the security of the data processing process.

[0021] To determine the axial conduction velocity component under standardized load excitation, the dynamic response stress measurement module executes a rate calculation program based on time-domain integration. The module acquires the velocity components after the subject inputs the standardized nutritional load. to The short-chain fatty acid content sequence within a time period was used to calculate the increase in the area under the curve (AUC) ΔA, and the time interval (Δt) during which the liver growth axis component rose from the baseline to 50% of its peak value was measured. The system then applied the formula... Calculate the axial conduction velocity component And using the calculated A linear step correction is performed on the transmission weight parameters in the growth steady-state evaluation module when the system detects... When the value is within the preset blockage threshold range, the gain ratio of the neuroendocrine component in the comprehensive evaluation is automatically increased. This dynamic mapping relationship based on energy flow fluctuations is used to identify latent developmental disorders in subjects that are masked by compensatory mechanisms in the resting state. Based on the transport lag in biochemical signal transmission between nodes of the gut-brain-liver axis, which causes a phase deviation between metabolic stimulation at the gut end and growth protein expression at the liver end on the time axis, the axial gating mapping module executes a phase alignment procedure based on physiological conduction delay. This module obtains the subject's age calibration factor σ, retrieves the biological index relationship model in the cloud database, establishes a mapping relationship between the age calibration factor σ and the baseline delay constant, calculates the ratio of the subject's current height value to the average standard height at the same developmental stage to determine the height deviation, and corrects the baseline delay constant based on the height deviation to generate a specific conduction delay τ. The specific conduction delay τ is determined by the system calling the inherent index relationship model of the cloud data synchronization module to analyze the intestinal metabolic rate of 5000 clinical samples. Cross-correlation analysis was performed on the Xie sequence and the liver growth factor response sequence to extract the maximum correlation lag steps of the transport time of biochemical substances in systemic and enterohepatic circulation. The product of the lag steps and the system sampling period was defined as the baseline time delay constant. A linear compensation term was introduced based on the subject's height deviation. For every 0.1 standard deviations from the standard mean of the same developmental stage, a 15-minute conduction time offset was added or subtracted to obtain the specific conduction delay τ used to perform asynchronous offset processing. The axial gating mapping module constructed a logical cache queue of length N in memory to store intestinal metabolic component data within continuous sampling periods. The intestinal metabolic component at time T-τ was retrieved by moving the logical pointer as the logical background of the liver growth axis component at time T. The module calculated the Euclidean distance of the short-chain fatty acid content at time T-τ relative to the subject's historical mean and converted it into a deviation ΔG. At the same time, the module extracted the first derivative of the neuroendocrine component at time T, filtered out the signal base noise generated by the neuronal activity inhibitory factor component, and obtained the fluctuation slope. The axial gating mapping module is based on the formula The gating calibration coefficient γ is calculated, where Φ and θ are preset response weight coefficients.

[0022] After obtaining the gating calibration coefficient γ to quantify the degree of axial conduction constraint, the growth steady-state evaluation module performs deep fusion of multi-source data. The module uses the gating calibration coefficient γ to adjust the response threshold of the liver growth factor receptor and performs nonlinear gain compensation on the liver growth axis component at time T, obtaining the gain-compensated liver growth axis component. The growth homeostasis evaluation module uses the following formula to calculate the development evaluation index. : In the formula, For the generated evaluation indicators, The liver growth axis components after gain compensation. γ represents the slope of the neuroendocrine component fluctuation, and γ is the gating calibration coefficient. , as well as The preset weighting factors, response weight coefficients Φ, θ, and weighting factors , , Based on adaptive determination of signal-to-noise ratio, the sample variance of each biochemical component of the subject within the preset monitoring period is calculated, the coefficient of variation (CV) reflecting the degree of data dispersion is determined, and the decision weights of each component are assigned as normalized values ​​inversely proportional to the CV. The weighting factor for the growth hormone-releasing hormone component with stable signal background is also calculated. Weighting factors for intestinal metabolic components affected by dietary and environmental noise, set between 0.4 and 0.6. The confidence level of the feedback calibration module is adjusted nonlinearly between 0.05 and 0.3 to ensure that the final output developmental evaluation index is accurate. Maintaining a confidence interval above 95%, the growth homeostasis evaluation module sends the calculation results to the display terminal, generating the subject's growth and development trajectory curve within a preset period and identifying its phase deviation relative to the standard growth curve, transforming the raw biochemical data into an intuitive assessment conclusion of developmental trends. To address the issue of intestinal noise caused by short-term dietary intake in complex clinical environments, the system includes a feedback calibration module for internal logic arbitration. This module uses developmental evaluation indicators as triggers to acquire the subject's hepatic bile acid metabolism characteristic data. Since bile acids are synthesized by the liver and secreted into the intestines, they have a regulatory effect on the intestinal microecology. The system establishes an expected coupling model between the fluctuation phase of bile acid components and intestinal characteristic data. The system performs confidence checks on the evaluation indicators through the fluctuation phase of bile acid metabolism characteristics. If the bile acid characteristics deviate from the expected interval derived from the liver output, the system determines that there is transient interference in the intestinal source. It reduces the weight of the intestinal metabolic component through a reliability weight lower limit locking logic, while increasing the decision weight of the neuroendocrine component, using the physiological feedback relationship between organs to ensure the consistency of the evaluation results.

[0023] The feedback calibration module establishes the reliability level of the intestinal signal source by executing a priori verification procedure based on hepatic metabolic signals. The system retrieves the correspondence records between bile acid component concentrations and intestinal microecological characteristic value sequences from a pre-stored large-scale clinical sample database, and uses a regression algorithm to extract an expected function reflecting the strength of inter-organ feedback inhibition. During real-time system operation, the module extrapolates the theoretical fluctuation range of intestinal characteristic values ​​based on the current bile acid excretion characteristics at the liver output end. If the measured short-chain fatty acid content deviates from the theoretical fluctuation range and its first derivative mutation frequency exceeds a preset threshold, the system determines that the intestinal signal source is interfered with by transient noise caused by dietary residues, and automatically triggers a reliability weight lower limit locking procedure to adjust the decision weight of intestinal metabolic components. Set the decision weight of the neuroendocrine component to 0.05. Compensation was applied to 0.65 to ensure the developmental assessment index. To maintain logical consistency under drastic environmental changes, and to address the pseudo-normal state of biochemical indicators at rest due to compensatory mechanisms, the system integrates a dynamic response stress testing module. This module acquires the growth slope of the intestinal metabolic component and the response delay of the liver growth axis component after the subject inputs a standardized nutritional load. The module calculates the product of the intestinal short-chain fatty acid ascent slope and the liver response rate to determine the axial conduction rate component, and corrects the conduction weight parameters in the growth homeostasis evaluation module accordingly. This stress testing logic identifies the dynamic bottleneck of the growth axis at the execution level by introducing excitation response analysis in the energy flow fluctuation process, providing quantitative indicators of latent axial conduction disorders for clinical use. The system retrieves the subject's historical evaluation trajectory through a cloud data synchronization module to ensure that the stored historical data is sequentially aligned with the current physiological components on the logical timeline.

[0024] Example 1: In the scenario of evaluating subjects whose biochemical signals fluctuate due to dietary cycles, the present invention provides a multidimensional data fusion evaluation of short stature development based on a quantitative model of the gut-brain-liver axis. The system collects heterogeneous biochemical data streams from subjects. Because the process by which short-chain fatty acids, metabolites from the gut, act on liver growth axis receptors involves time-consuming systemic transport, there is a non-instantaneous spatiotemporal shift between the excitation signal peak at the gut end and the growth protein response peak at the liver end. The linear weighting performed by traditional systems under the same sampling period can lead to a phase misalignment of causal logic, making the system unable to identify the risk of slowed growth rate caused by feedback inhibition. For this situation, the system's data acquisition module acquires in vitro sample detection data through a standardized application programming interface, including the short-chain fatty acid content measured by gas chromatography-mass spectrometry and the content measured by chemiluminescence immunoassay. The module collects data on the growth hormone-releasing hormone (GH) component and serum insulin-like growth factor-I (IGF-I) concentration. After receiving the data, it uses a hash algorithm to perform de-identification encryption, converting the subject's identity label into a desensitized identifier. The axial gating mapping module calculates the subject's age calibration factor σ and corrects the baseline time delay constant based on height deviation, determining the specific conduction delay τ for the subject to be 4.5 hours. This module establishes a pre-defined logical cache queue of length N in memory to store intestinal metabolic data over consecutive periods. By backtracking the sampling pointer by 4.5 hours, it retrieves the short-chain fatty acid (SCFA) content data at time T-τ as the parameter background for the liver growth axis component at the current time T. The module calculates the deviation ΔG of the SCFA content at time T-τ relative to the subject's historical mean, and simultaneously extracts the first derivative of the GH component at time T to determine the fluctuation slope. The axial gating mapping module calculates the asynchronously aligned gating calibration coefficients according to the following formula. : Where γ is the gating calibration coefficient, Φ and θ are preset response weighting coefficients, and ΔG is the deflection steering magnitude. To address the fluctuation slope, this procedure establishes a time-series asynchronous offset processing logic, enabling the originally misaligned excitation and response signals to achieve causal alignment on the logical time axis, thus eliminating model drift caused by the asynchronous nature of signal flow.

[0025] The growth homeostasis evaluation module receives the gating calibration coefficient γ, performs weighted gain adjustment on the insulin-like growth factor-I concentration collected at time T, and calculates the liver growth axis components after gain compensation. The system ultimately calculates and outputs developmental evaluation indicators based on the following formula. : In the formula, As a developmental evaluation indicator, The liver growth axis components after gain compensation. Where γ is the fluctuation slope and γ is the gating calibration coefficient. , as well as The pre-defined weighting factor; in this calculation path, the gating calibration coefficient γ acts as a dynamic logic valve, incorporating variables reflecting the strength of intestinal negative feedback into the evaluation weights of growth proteins, thus affecting the final output index. It can reflect the systematic homeostasis of the subject's growth and development. The system finally generates a quantitative evaluation report that reflects the subject's growth and development trajectory, including the subject's growth and development homeostasis score within a preset period, and displays the phase deviation value of the growth and development trajectory curve relative to the standard growth curve.

[0026] Example 2: On a hardware simulation platform integrating a high-throughput in vitro biochemical analysis interface, this experimental group verified the steady-state determination efficiency of a multidimensional data fusion evaluation system for short stature development based on a gut-brain-liver axis quantification model when processing high-noise, asynchronous biochemical signals. The data used in the experiment came from clinically desensitized in vitro serum samples and intestinal lavage fluid sequencing data. The data acquisition module was required to have an analog-to-digital conversion accuracy of no less than 12 bits, and the sampling frequency was set to 10Hz to cover the transient fluctuation characteristics of biochemical indicators. The setting of the sampling period balanced the integrity of biochemical signal capture with the computational load of the system. When the effective frequency bandwidth of the monitored signal was in the range of 0.1Hz to 1Hz, the sampling frequency was increased to 10 times its upper limit to avoid signal aliasing under the Nyquist sampling theorem. The experimental environment simulated clinical interference, with random perturbations with a signal-to-noise ratio of 20dB superimposed on the signal source to simulate the irregular metabolic noise of the subjects during their dietary cycle.

[0027] The sample group of this invention operates on the gut-brain-liver axis quantification logic, setting the age calibration factor σ of the subjects to 0.85 and the height deviation to 0.92. The specific conduction delay τ calculated by the axial gating mapping module is 4.2 h. In contrast, the control group A uses a linear weighting method and does not run sequential asynchronous offset processing, i.e., τ is set to 0. After inputting a sample sequence with an average original short-chain fatty acid content of 5.2 mmol / L and an average insulin-like growth factor-I (IGF-I) concentration of 115 ng / mL, the sample group of this invention backtracks 4.2 h of gut background data through a logic cache queue, generating a gating calibration coefficient γ with an average value of 0.78. Experimental data show that within the non-steady-state range triggered by dietary fluctuations, the developmental evaluation indicators output by the sample group of this invention... The standard deviation of the input was 0.045, while the standard deviation of the control group A under the same input reached 0.185. Because the axial gating mapping module established spatiotemporal phase alignment, the originally generated logical drift was canceled out during the calculation process, confirming the contribution of asynchronous fusion logic to model stability. To demonstrate the synergistic effect of the various functional units of the system, a control group B was further established. This group only retained the coupling logic between the neuroendocrine component and the liver growth axis component, removing the feedback input of the intestinal metabolic component. Under the scenario of enhanced feedback inhibition generated by the simulated intestinal microbial community, the deviation vector ΔG of the core problem variable, namely short-chain fatty acids, increased from 0.1 to 0.5. Data examples show that with the increase of ΔG, the developmental evaluation index of the sample group of this invention... It exhibited a monotonically decreasing trend consistent with clinical growth deceleration observations, with an index sensitivity slope of 0.65; while the evaluation index of control group B... The value remained fluctuating around 0.95, and growth arrest caused by gut-derived feedback could not be identified. The nonlinear interaction between the gut metabolic component and the neuroendocrine component through the gating calibration coefficient γ was confirmed, providing the system with a causal evaluation dimension that cannot be covered by a single endocrine indicator.

[0028] In the empirical verification of key parameter boundaries, the system performed out-of-range stress tests on the gating weight parameters Φ and θ, with its operating range limited to 0.1 to 0.9. When the set value of Φ exceeded the upper limit by 20%, the amplification factor of intestinal noise increased from 1.2 times to 3.5 times, leading to changes in developmental evaluation indicators. When saturation occurs, the growth rate slows down, and it becomes impossible to distinguish between pathological blockage and physiological fluctuations; when Φ is below the lower limit of 50%, the system's sensitivity loss rate to feedback inhibition signals reaches 72%; the system uses an internal logic arbitration through a feedback calibration module to cope with dietary interference, and when the transient peak value of simulated intestinal metabolites exceeds 300% of the historical average, the system extracts the fluctuation phase of bile acid metabolism characteristics; the experiment observed that the sample group of this invention automatically assigns decision weights to intestinal metabolic components. The weight of the neuroendocrine component was reduced from 0.3 to 0.05. Increased to 0.65; developmental evaluation index after treatment The confidence interval recovered to over 95%; the final output developmental evaluation index The correlation coefficient with the subjects' actual protein synthesis rate increased from 0.68 to 0.91.

[0029] Example 3: This example combines Figures 1 to 2 This section describes a multidimensional data fusion evaluation system for short stature development based on a quantitative model of the gut-brain-liver axis. Figure 1As shown, the data processing flow begins with the parallel acquisition and time-series processing of multi-source signals. The intestinal metabolic component module acquires the raw metabolic data at time T, and the data stream enters the asynchronous offset processing stage to retrieve... The time-time parameter benchmark is used, while the neuroendocrine component module is responsible for acquiring endocrine data at time T and calculating the rate of change of the neural component through the fluctuation slope extraction step. After the above two data branches converge, the gating calibration coefficient γ is calculated based on the deviation directional amount and slope. It is transmitted to the execution gain compensation step as an adjustment command. In this step, the system combines the liver growth axis component of the acquired growth axis data at time T and adjusts the liver growth factor response threshold. Finally, the data is fused to generate growth and development evaluation indicators and output steady-state evaluation results.

[0030] like Figure 2 As shown, a fishbone diagram architecture is used to illustrate the support relationship between each functional module and the technical goal. The data acquisition module in the upper left corner covers three dimensions: intestinal metabolic components, liver growth axis components, and neuroendocrine components. The axial gating mapping module in the upper right corner includes functions such as gating calibration coefficient calculation, fluctuation slope characteristic determination, and dynamic retrieval of parameter benchmarks. The signal timing logic branch in the lower left corner involves phase misalignment elimination, asynchronous offset processing, and determination of conduction delay parameters. The growth steady-state evaluation module branch in the lower right corner performs multi-dimensional data fusion, evaluation model drift elimination, and liver growth axis gain compensation. All of the above functional branches converge towards the final goal at the head of the fishbone diagram, which is to achieve accurate output of growth and development evaluation indicators.

[0031] Example 4: In a short stature development evaluation system with multiple concurrent in vitro sampling interfaces, the data acquisition module receives 16-bit resolution biochemical data streams via a universal serial bus hardware port. The sampling frequency is set to 1Hz. The system's main processor allocates a 2048MB static random access memory space as a logical cache queue to store complete biochemical characteristic sequences from the past 24 hours. The system identifies the starting point of metabolic fluctuations using a peak detection operator based on comparator logic, and uses 120 minutes after the starting point as the effective time window for calculating developmental evaluation indicators. All input signals are processed within the specified time frame. Before entering the core logic, all data must pass through a digital low-pass filter with a cutoff frequency of 0.5Hz to filter out power frequency interference and high-frequency glitches during the in vitro sampling process, ensuring the background purity of the data involved in the deviation traverse calculation. The system's main processor is allocated 2GB of high-speed cache memory to store the logic cache queue data. The system runs the calibration procedure of the biological index relationship model, retrieving a pre-stored sample set of 5000 sets containing age, height, and measured cycle time, and uses the least squares method to calculate the time delay constant coefficient matrix. The biological index relationship model uses the subject's age calibration factor σ and height deviation as input vectors. Through calculation The conduction delay τ is determined by multiplying the delay constant coefficient matrix; the axial gating mapping module locates the starting address of the data block in the cache memory based on the conduction delay τ, obtains the intestinal metabolic component in the storage unit corresponding to time T-τ by moving the read pointer, and compares the intestinal metabolic component with the fluctuation slope of the neuroendocrine component at time T. A coupling operation is performed to determine the gating calibration coefficient γ, which reflects the strength of the gut-brain feedback; where σ is the age calibration factor. Let τ be the input vector, τ be the propagation delay in hours (h), and T be the current sampling time. γ represents the slope of the neuroendocrine component fluctuation, and γ is the gating calibration coefficient.

[0032] The growth homeostasis evaluation module executes a parameter adaptive procedure based on signal quality. The system selects a sliding window of 60 seconds within the sampling period and calculates the real-time variance of the intestinal metabolic component and the neuroendocrine component within the sliding window. and The dynamic weighting factor of the components is determined according to the following formula. : ,in, Let K be the dynamic weighting factor for the i-th component, and K be a preset normalization constant. This represents the variance of the corresponding component. This represents the real-time variance of intestinal metabolic components. This represents the real-time variance of the neuroendocrine component; when the intestinal metabolic component is affected by dietary residue, causing a jump, it leads to... When the factor is increased by 5 times, the system will dynamically adjust the weighting factor. The value was lowered from 0.3 to 0.06, utilizing the inverse proportional adjustment mechanism of the dynamic weighting factor to offset the interference of external disturbances on the growth homeostasis determination results, thus ensuring the development evaluation index... Maintain a confidence interval above 95%; where, These are indicators for developmental evaluation.

[0033] Example 5: In data processing scenarios involving multi-center clinical collaboration, the system runs offline data filling and dynamic model maintenance procedures for the biological index relationship model; by accessing the cloud database to extract a dataset containing 50,000 in vitro sample detection records, the subject's age calibration factor σ and height deviation are mapped to a multi-dimensional feature space, and a nonlinear regression algorithm is used to perform weight updates on the time delay constant coefficient matrix; within the system initialization period, the module calculates the temporal cross-correlation of historical sampling sequences, determines the physical time distribution of each biochemical component in the systemic circulation transport process, and adjusts the prediction model parameters of a specific conduction delay τ based on the calculated residual mean, so that the system maintains the causal alignment of the evaluation logic when facing subject groups from different geographical regions.

[0034] When the system is deployed in a clinical testing environment with varying levels of precision in in vitro biochemical analysis, it executes an on-site baseline calibration procedure after connecting to an external sensor interface. This requires the data acquisition module to obtain baseline values ​​of biochemical components from the subject at rest and to calculate the real-time variance of short-chain fatty acid content. Determine the current environmental biochemical background noise and apply it to the weighting factors in the growth steady-state evaluation module. The initial step size is corrected; if the data sensing unit detects that the ratio of the sampling frequency to the biochemical substance response period is less than 5, the system automatically starts the interpolation compensation operator to enhance the smoothness of the time-domain signal, and works with the dynamic response pressure testing module to perform initial offset measurement on the axial conduction velocity component under standardized nutrient load. A standardized pre-debugging procedure is used to address the growth and development evaluation indicators caused by the heterogeneity of the deployment environment. Addressing drift issues and improving system stability during long-term monitoring.

[0035] Example 6: In a pediatric collaborative network integrating in vitro biochemical sampling terminals of varying precision, the system runs a baseline calibration procedure with multi-source data streams. Fasting resting-state data streams from subjects are collected via a serial bus port during the 2-hour pre-deployment phase, and the sample variance of short-chain fatty acid content is calculated. The system determines the initial deviation by combining the analog-to-digital conversion resolution of the detection equipment, and then performs weight fine-tuning on the time delay constant coefficient matrix in the axial gating mapping module. During this process, the system uses a correlation coefficient algorithm to identify the hysteresis order between the intestinal metabolic fluctuation edge and the neuroendocrine signal response edge. The product of the calculated hysteresis steps and the sampling period is used as the calibration increment for a specific conduction delay τ to offset the phase deviation caused by inconsistent sampling frequencies. Let τ be the sample variance, and τ be the specific propagation delay in hours.

[0036] The system runs a dimensional alignment procedure. Before performing multi-dimensional data overlay, the growth homeostasis evaluation module performs normalization processing on each component based on range transformation to obtain the developmental baseline sequence of the same sex population at the current developmental stage of the subject, and determines the maximum observed value of biochemical indicators. With minimum observed value And calculate the component values ​​according to the following formula. : In the formula, Here are the normalized component values, and X is the original input parameter. For the maximum observed value, Minimum observed value; processed liver growth axis components Slope of fluctuation of neuroendocrine component Mapping is performed within the same numerical range. When the data acquisition module detects that the voltage amplitude of the sampled signal is below 0.5mV for 10 seconds or a feedback communication interruption signal is detected, the system initiates a linear interpolation compensation program based on the historical evolution slope, using a baseline alignment program to offset the developmental evaluation indicators caused by sensor temperature drift. The numerical fluctuations, among which, For the liver growth axis components, The slope of the fluctuation. These are developmental evaluation indicators; the system stores normalized developmental evaluation indicators. The system generates a growth steady-state assessment report at the terminal and identifies the deviation of the evaluation index curve from the standard curve. By executing a normalization mapping mechanism and an interpolation compensation procedure for abnormal fluctuations in ex vivo samples, the system converts the cross-organ feedback process into a data stream supported by a defined range operator and a linear regression operator, thus completing the assessment of the causes of developmental arrest.

[0037] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0038] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A multidimensional data fusion evaluation system for short stature development based on a quantitative model of the gut-brain-liver axis, characterized in that, The system includes a data acquisition module, an axial gating mapping module, and a growth steady-state evaluation module. The data acquisition module is used to acquire the subject's intestinal metabolic components, liver growth axis components, and neuroendocrine components; The axial gating mapping module, connected to the data acquisition module, is used to determine the transmission delay τ of intestinal metabolic signals to the liver and brain based on the subject's physiological characteristics. The axial gating mapping module also performs asynchronous offset processing, retrieving the intestinal metabolic component at time T-τ as the parameter baseline for the current time T, and calculating the deviation ΔG of the intestinal metabolic component at time T-τ relative to the subject's historical mean. Furthermore, the axial gating mapping module extracts the oscillation slope of the neuroendocrine component at time T. And based on the deflection ΔG and the oscillation slope And the gate calibration coefficient is calculated using the preset response weighting coefficients Φ and θ. ; The growth steady-state evaluation module, connected to the axial gating mapping module, is used to perform gain compensation on the liver growth axis components at time T according to the gating calibration coefficient γ, and to evaluate the liver growth axis components after gain compensation. Data is fused with neuroendocrine components to generate and output evaluation indicators characterizing the homeostasis of the subject's growth and development.

2. The multidimensional data fusion evaluation system for short stature development based on a gut-brain-liver axis quantitative model according to claim 1, characterized in that, The axial gating mapping module determines the conduction delay τ by following these steps: Step S1, obtain the subject's age calibration factor σ; Step S2, retrieve the biological index relationship model and establish a mapping relationship between the age calibration factor σ and the baseline delay constant; Step S3, correct the baseline delay constant based on the subject's height deviation to generate the conduction delay τ.

3. The multidimensional data fusion evaluation system for short stature development based on a quantitative model of the gut-brain-liver axis as described in claim 1, characterized in that, When the axial gating mapping module extracts the deflection ΔG, it is used to calculate the Euclidean distance between the current short-chain fatty acid content and the mean level of subjects at the same developmental stage in the database, and convert the Euclidean distance into the deflection ΔG; among which, the intestinal metabolic component includes the short-chain fatty acid content and the intestinal microecological characteristic value sequence.

4. The multidimensional data fusion evaluation system for short stature development based on a quantitative model of the gut-brain-liver axis as described in claim 1, characterized in that, The system also includes a feedback calibration module; the feedback calibration module is connected to the growth homeostasis evaluation module and is used to obtain the bile acid metabolism characteristics of the subjects as the trigger source and use them as internal arbitration signals, and to perform confidence verification on the evaluation indicators according to the fluctuation phase of the bile acid metabolism characteristics.

5. The multidimensional data fusion evaluation system for short stature development based on a gut-brain-liver axis quantitative model according to claim 1, characterized in that, The system also includes a dynamic response stress test module; the dynamic response stress test module is used to obtain the growth slope of the intestinal metabolic component and the response delay of the liver growth axis component after the subject inputs a standardized nutritional load, and to correct the transmission weight parameters in the growth homeostasis evaluation module accordingly.

6. The multidimensional data fusion evaluation system for short stature development based on a quantitative model of the gut-brain-liver axis as described in claim 1, characterized in that, When the growth steady-state evaluation module performs gain compensation, it is used to determine the gating calibration coefficient. Adjust the response threshold of liver growth factor receptor.

7. The multidimensional data fusion evaluation system for short stature development based on a gut-brain-liver axis quantitative model according to claim 1, characterized in that, The data acquisition module is also connected to the cloud data synchronization module; the cloud data synchronization module is used to retrieve the subject's historical evaluation trajectory and align the execution sequence of each physiological component with the intestinal stimulation characteristics in the historical spatiotemporal context.

8. The multidimensional data fusion evaluation system for short stature development based on a gut-brain-liver axis quantitative model according to claim 1, characterized in that, The neuroendocrine component includes the growth hormone-releasing hormone component and the neuronal activity inhibitory factor component; the axial gating mapping module calculates the fluctuation slope. At time T, the first derivative of the growth hormone-releasing hormone component at time T is extracted, and the signal base noise generated by the neuronal activity inhibitory factor component is filtered out.

9. A multidimensional data fusion evaluation system for short stature development based on a quantitative model of the gut-brain-liver axis, as described in claim 1, is characterized in that... The growth steady-state evaluation module uses the following formula to calculate the evaluation index. : ,in, For the generated evaluation indicators, This refers to the liver growth axis components after gain compensation. The neuroendocrine component is represented by γ, which is the gating calibration coefficient. , and This is a preset weighting factor.

10. A multidimensional data fusion evaluation system for short stature development based on a gut-brain-liver axis quantitative model according to claim 1, characterized in that, The system also includes a display terminal; the display terminal is connected to the growth steady-state evaluation module, which is used to generate the growth and development trajectory curve of the subject within a preset period based on the evaluation indicators, and to identify the phase deviation of the growth and development trajectory curve relative to the standard growth curve.

Images