Excrement weight change and panda estrus cycle correlation modeling method and system

Abstract

The invention provides an excrement weight change and panda oestrus cycle correlation modeling method and system, and relates to the technical field of panda breeding, and the method comprises the steps: obtaining basic monitoring data of female panda individuals; performing fiber metabolism characteristic extraction according to the daily excrement weighing value to obtain a bamboo fiber metabolism efficiency sequence; performing matrix construction according to the metabolic efficiency sequence and the serum progesterone concentration detection value to obtain a progesterone-metabolic coupling matrix; correcting according to the natural season classification label to obtain an inhibition factor; identifying according to the progesterone-metabolism coupling matrix and the inhibition factor to obtain an estrus critical state characteristic value; and performing model construction according to the metabolic efficiency sequence, the progesterone-metabolic coupling matrix, the inhibition factor and the estrus critical state characteristic value to obtain a panda estrus window dynamic prediction model. According to the method, the missing rate of artificial insemination window identification is reduced, and the breeding efficiency of endangered species is greatly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of giant panda breeding. Specifically, it relates to a method and system for establishing a correlation model between fecal weight changes and the estrous cycle of giant pandas. Background Art

[0002] In the field of endangered species conservation, the core problem in the artificial breeding of giant pandas focuses on the accurate prediction of their estrus window period - this physiological stage has a very short duration (usually only 24 - 72 hours), and shows significant individual differences and seasonal fluctuation patterns. The existing technical means mainly rely on two types of methods: Interventional serum hormone monitoring requires frequent blood sampling to track changes in progesterone concentration, but the sampling process is extremely likely to cause stress reactions, resulting in abnormal elevation of cortisol levels and interfering with reproductive endocrine homeostasis; The ethological observation method makes artificial judgments by identifying estrus behaviors such as rubbing the genitals and special calls, but the early behavioral signals are highly concealed and are easily interfered by factors such as noise in the captive environment and differences in the experience of the keepers, with a very high misjudgment rate. Although the recently emerging non-invasive fecal monitoring technology avoids the risk of animal stress, its existing system still has key bottlenecks: the lack of a correlation mechanism between fecal weight and metabolic efficiency, the lack of a dynamic coupling model for multiple physiological signals, and the inability to capture the critical transient characteristics of estrus, resulting in insufficient prediction accuracy and the urgent need to improve the efficiency of artificial breeding.

[0003] Based on the above-mentioned disadvantages of the existing technology, there is an urgent need for a method and system for establishing a correlation model between fecal weight changes and the estrus cycle of giant pandas. Summary of the Invention

[0004] The purpose of the present invention is to provide a method for establishing a correlation model between fecal weight changes and the estrus cycle of giant pandas to improve the above problems. To achieve the above purpose, the technical solutions adopted by the present invention are as follows: In the first aspect, the present application provides a method for establishing a correlation model between fecal weight changes and the estrus cycle of giant pandas, including: Obtaining the basic monitoring data of female giant panda individuals, including daily fecal weight measurement values, serum progesterone concentration detection values, and natural season classification labels; Extracting fiber metabolism characteristics based on the daily fecal weight measurement values to obtain a metabolic efficiency sequence of bamboo fibers; Constructing a matrix based on the metabolic efficiency sequence and the serum progesterone concentration detection values to obtain a progesterone-metabolism coupling matrix; Making corrections according to the natural season classification labels, and obtaining an inhibition factor by quantifying the inhibitory effect of seasonal fluctuations of bamboo leaf secondary metabolites on the fermentation rate; Identifying based on the progesterone-metabolism coupling matrix and the inhibition factor, and obtaining an estrus critical state characteristic value by detecting the derivative jump characteristics and stability characteristics of the corrected coupling matrix within a preset time window; Based on the metabolic efficiency sequence, the progesterone-metabolism coupling matrix, the inhibitory factor, and the estrus critical state eigenvalue, a model is constructed to obtain a dynamic prediction model for the estrus window of giant pandas.

[0005] In a second aspect, the present application also provides a system for associating the change in fecal weight with the estrus cycle of giant pandas, including: An acquisition module for acquiring basic monitoring data of female giant panda individuals, including daily fecal weight measurement values, serum progesterone concentration detection values, and natural season classification labels; An extraction module for extracting fiber metabolism characteristics based on the daily fecal weight measurement values to obtain a metabolic efficiency sequence of bamboo fibers; A construction module for constructing a matrix based on the metabolic efficiency sequence and the serum progesterone concentration detection values to obtain a progesterone-metabolism coupling matrix; A correction module for correction according to the natural season classification labels, and obtaining an inhibitory factor by quantifying the inhibitory effect of the seasonal fluctuation of bamboo leaf secondary metabolites on the fermentation rate; An identification module for identification based on the progesterone-metabolism coupling matrix and the inhibitory factor, and obtaining an estrus critical state eigenvalue by detecting the derivative jump characteristics and stability characteristics of the corrected coupling matrix within a preset time window; An output module for constructing a model based on the metabolic efficiency sequence, the progesterone-metabolism coupling matrix, the inhibitory factor, and the estrus critical state eigenvalue to obtain a dynamic prediction model for the estrus window of giant pandas.

[0006] The beneficial effects of the present invention are as follows: By establishing a quantitative characterization mechanism for the metabolic efficiency sequence of bamboo fibers, the interference of feeding fluctuations is stripped, and the mapping accuracy between fecal weight changes and core metabolic characteristics is significantly improved; by constructing a progesterone-metabolism coupling matrix and introducing seasonal inhibitory factors for dynamic compensation, multi-dimensional collaborative modeling of the digestive system rhythm and reproductive endocrine signals is realized, and the problem of coupling breakage in the physiological system is solved; based on the dual-characteristic spatio-temporal detection mechanism of the estrus critical state, the transient correlation characteristics of the system stability jump and the hormone flux derivative step are captured, and the estrus critical point is accurately located; finally, through the time-domain alignment and dynamic weight fusion of multi-source physiological signals, a high-confidence probability prediction model for the estrus window is formed, reducing the missed rate of identifying the artificial insemination window and greatly improving the breeding efficiency of endangered species. Description of the Drawings

[0007] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as limiting the scope. For those of ordinary skill in the art, without creative efforts, other related drawings can also be obtained based on these drawings.

[0008] Figure 1 It is a schematic flowchart of a method for establishing a correlation model between fecal weight change and the estrus cycle of giant pandas in the embodiments of the present invention; Figure 2 It is a schematic structural diagram of a system for establishing a correlation model between fecal weight change and the estrus cycle of giant pandas in the embodiments of the present invention; Figure 3 It is a schematic structural diagram of a device for establishing a correlation model between fecal weight change and the estrus cycle of giant pandas in the embodiments of the present invention; Figure 4 It is a flowchart for constructing an estrus state distribution model in the embodiments of the present invention.

[0009] Reference numerals in the figure: 800, a device for establishing a correlation model between fecal weight change and the estrus cycle of giant pandas; 801, a processor; 802, a memory; 803, a multimedia component; 804, an I / O interface; 805, a communication component; 901, an acquisition module; 902, an extraction module; 903, a construction module; 904, a correction module; 905, an identification module; 906, an output module. Detailed implementation manners

[0010] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Usually, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed present invention, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0011] It should be noted that similar reference numerals and letters indicate similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. At the same time, in the description of the present invention, terms such as "first" and "second" are only used for differential description and cannot be understood as indicating or implying relative importance.

[0012] Example 1:

[0013] This example provides a method for modeling the association between fecal weight changes and the estrous cycle of giant pandas.

[0014] Refer to Figure 1 , which shows that this method includes steps S100 to S600.

[0015] Step S100: Obtain the basic monitoring data of female giant panda individuals, including daily fecal weight measurement values, serum progesterone concentration detection values, and natural season classification labels; In this step, three key physiological data of female giant pandas are obtained: The daily fecal weight measurement value reflects the quality change in the process of bamboo fiber digestion and is a direct observation index of the output of the digestive system; the serum progesterone concentration detection value characterizes the periodic fluctuation of reproductive hormones and is the core biomarker of the estrus state; the natural season classification label encodes environmental factors (spring / summer / autumn / winter) and is used to quantify the seasonal differences in the chemical composition of bamboo plants. This step establishes a multi-dimensional physiological monitoring basis and circumvents the limitations of a single data source.

[0016] Step S200: Extract fiber metabolism characteristics based on the daily fecal weight measurement values to obtain a metabolic efficiency sequence of bamboo fiber; In this step, by separating the feeding behavior noise in the fecal weight (such as abnormal weight peaks caused by overeating in winter), the basic metabolic component that only reflects the digestion efficiency of bamboo fiber is extracted; combined with the analysis of colonic rhythm characteristics, the intensity of cecal microbial fermentation is captured from the biophysical level, and finally a metabolic efficiency sequence of bamboo fiber is generated. This step realizes the transformation from "weight signal" to "metabolic essence" and breaks through the bottleneck that traditional fecal indicators cannot distinguish physiological processes from behavioral interference.

[0017] Step S300: Construct a matrix based on the metabolic efficiency sequence and the serum progesterone concentration detection value to obtain a progesterone-metabolism coupling matrix; Preferably, in this step, a dynamic response analysis is performed on the metabolic efficiency sequence and the progesterone concentration, and it is found that a hysteresis law of periodic fluctuation of metabolic efficiency appears 72 hours after the change in progesterone concentration. By constructing a three-dimensional response surface representing the hormone-metabolism dose relationship, where the X-axis represents the progesterone concentration, the Y-axis represents the time delay, and the Z-axis represents the metabolic gain, a progesterone-metabolism coupling matrix is formed, and a quantitative association model between the digestive system and the reproductive system is established.

[0018] Step S400: Make corrections according to the natural season classification label, and obtain an inhibition factor by quantifying the inhibitory effect of seasonal fluctuations of bamboo leaf secondary metabolites on the fermentation rate; This step analyzes the periodic fluctuation law of bamboo secondary metabolites based on natural season classification tags. By quantifying the dose-dependent inhibitory effect of this substance on cecal cellulase activity, the biochemical inhibitory effect is converted into a correction parameter for the coupling matrix - the inhibition factor. The core of this step is to establish a quantitative mapping between environmental season variables and digestive physiological disturbances, and solve the problem of model prediction deviation caused by seasonal variation of bamboo-derived phytochemical components.

[0019] Step S500: Identify according to the progesterone-metabolism coupling matrix and the inhibition factor. By detecting the derivative jump characteristics and stability characteristics of the corrected coupling matrix within a preset time window, the estrus critical state eigenvalue is obtained. Preferably, in the season-corrected coupling matrix, by detecting the spatio-temporal coupling event of the first-order derivative jump of hormone flux within a 48-hour window (mathematical representation of hypothalamic GnRH pulse burst) and the cliff-like decline of system stability (Lyapunov index < 0.15 for 6 hours as a marker of ovarian feedback activation), a binary estrus critical state eigenvalue is generated. This step realizes the accurate conversion from continuous physiological signals to discrete ovulation events.

[0020] Step S600: Construct a model according to the metabolic efficiency sequence, progesterone-metabolism coupling matrix, inhibition factor and estrus critical state eigenvalue to obtain a dynamic prediction model for the estrus window of giant pandas.

[0021] It can be understood that this step eliminates the time difference in multi-source signal acquisition through the time axis alignment technology, assigns different contribution degrees to the metabolic basic characteristics, hormone response intensity, inhibition factor and critical signal, and constructs a weighted feature vector; and uses the sliding time window mechanism to continuously output the future estrus probability, forming a dynamic prediction model for the estrus window. Finally, this model takes the change in fecal weight as the input starting point, and through analyzing the metabolic efficiency of bamboo fiber, integrating hormone response and season interference correction, and capturing the reproductive critical transient characteristics, finally outputs a function mapping system with a high-precision probability distribution of the future estrus window. It realizes the direct mapping from fecal weight change to estrus cycle probability, providing a quantitative decision-making basis for artificial breeding.

[0022] Furthermore, step S200 includes steps S210 to S230.

[0023] Step S210: Separate the feeding fluctuation according to the daily fecal weight measurement value. By decomposing the moving standard deviation in the time domain, the abnormal fluctuation component of feeding frequency is separated to obtain the basic metabolism component sequence. Specifically, this step uses the moving standard deviation time domain decomposition technology to strip the abnormal weight fluctuation caused by the sudden change in feeding frequency (such as overeating in winter), and retains the component sequence reflecting the basic metabolism steady state. The core is to solve the physiological paradox of "high food intake ≠ high metabolic efficiency" and extract the basic metabolism component sequence synchronized with the digestion cycle.

[0024] Step S220: Extract the colon rhythm based on the basal metabolic component sequence. Extract the energy value of the characteristic frequency band of cecal cellulose fermentation through wavelet packet analysis to obtain the colon rhythm energy sequence. Preferably, in this step, based on the wavelet packet energy focusing principle, the 2-4 Hz characteristic frequency band (corresponding to the cecal chamber resonance frequency) is separated from the basal metabolic components to quantify the cellulose fermentation intensity. By capturing the fermentation rhythm biomarker generated by the unique colon anatomical structure of the giant panda, the colon rhythm energy sequence is output.

[0025] Step S230: Integrate and process the metabolic efficiency according to the basal metabolic component sequence and the colon rhythm energy sequence. By analyzing the synergistic mechanism between colon motility and steady-state metabolism and correcting the influence of the physical property variation of biological samples, the bamboo fiber metabolic efficiency sequence is obtained.

[0026] It can be understood that in this step, the biological synergistic effect between colon mechanical peristalsis (basal metabolic component) and microbial fermentation (rhythm energy) is analyzed. By correcting the physical interference of fecal moisture content variation on the weight signal, the bamboo fiber metabolic efficiency sequence is generated. This process deconstructs the fecal weight value vulnerable to environmental interference into the characteristics of three biological processes reflecting the unique digestive physiology of the giant panda, and finally obtains the bamboo fiber metabolic efficiency sequence that is stripped of the interference of feeding behavior, seasonal moisture, and physical properties.

[0027] Furthermore, step S300 includes steps S310 to S330.

[0028] Step S310: Quantify the metabolic fluctuation pattern of the digestive tract metabolism according to the bamboo fiber metabolic efficiency sequence. By extracting the periodic characteristics of the metabolic efficiency and the precursor signal of the critical point, the reproductive cycle metabolic characteristic sequence is obtained. It should be noted that in this step, the periodic fluctuation pattern of the bamboo fiber metabolic efficiency sequence is recognized, and the characteristic steep rise and fall turning pattern is extracted through time window analysis. The sliding peak-valley detection method is used to capture the stage transition characteristics of the metabolic efficiency near the estrus period, and the synchronous relationship between digestive metabolism fluctuation and reproductive cycle is identified. The continuous metabolic fluctuation is quantified into a discrete sequence representing the evolution of the reproductive stage, establishing a bridge from digestive tract metabolic activities to reproductive physiological states.

[0029] Step S320: Evaluate the hormone time-delay effect according to the detected value of serum progesterone concentration. By identifying the delayed regulation law of the progesterone concentration gradient change on intestinal function, the progesterone regulation kinetic sequence is obtained. Based on the measured value of serum progesterone concentration, lagged correlation analysis is performed in this step. By defining a fixed delay window to track the correspondence between the change in progesterone concentration and the metabolic response, a mapping model of the magnitude of hormone change and the intestinal function response is constructed. A saturation function is used to fit the non-linear effect of the progesterone concentration gradient on metabolic regulation, forming a kinetic sequence that quantifies the intensity of hormone regulation, and revealing the law of delayed regulation of reproductive endocrinology on the digestive system.

[0030] Step S330: Based on the metabolic characteristic sequence of the reproductive cycle and the progesterone regulation kinetic sequence, a biological coupling relationship model is established. By constructing a hormone-metabolism dose-response surface to characterize the three-dimensional association, a progesterone-metabolism coupling matrix is obtained.

[0031] It can be understood that in this step, a spatio-temporal association model is established between the metabolic characteristic sequence and the hormone kinetic sequence. By establishing a three-dimensional coordinate system and using surface interpolation technology to generate a continuous response surface. Feature point enhancement and encryption are implemented in the progesterone peak region and the time-delay sensitive region, and finally a quantified progesterone-metabolism coupling matrix is output, realizing the modeling conversion from a linear relationship to a spatial association.

[0032] Furthermore, step S400 includes steps S410 to S430.

[0033] Step S410: Based on the natural season classification label, a dynamic analysis of the secondary substances of bamboo plants is carried out. By establishing a mapping relationship between the season cycle and the fluctuation of the leaf tannic acid concentration, a concentration characteristic sequence of secondary metabolites is obtained; It should be noted that in this step, a mapping model between the natural season classification label and the concentration of bamboo leaf secondary metabolites is established, and the season characteristics are transformed into a concentration sequence of secondary substances. Based on the physiological laws of bamboo plants confirmed by the literature, the seasonal fluctuation trend of secondary metabolites (such as tannins) is quantified, and a concentration characteristic sequence that strictly corresponds to the time dimension and the season label is output. This step avoids the operation complexity of direct chemical detection and realizes the quantitative conversion from environmental factors to biochemical interference.

[0034] Step S420: Based on the concentration characteristic sequence of secondary metabolites, an evaluation and treatment of the digestive inhibition effect is carried out. By analyzing the dose inhibition effect of tannins on the cellulase activity of cecal microorganisms, a fermentation inhibition intensity sequence is obtained; Based on the concentration sequence of secondary substances, a dose-inhibition effect response model is constructed in this step. By analyzing the negative correlation between the concentration of specific secondary substances and the cellulase activity, a non-linear function is used to characterize the biological law of "weak inhibition in the low concentration range and a sharp increase in the inhibition rate after the high concentration threshold". The output fermentation inhibition intensity sequence directly quantifies the dynamic inhibition intensity of the bamboo leaf chemical components on the fermentation efficiency of cecal microorganisms, and each value in this sequence represents the theoretical inhibition rate of the secondary substances on the digestive efficiency on the day.

[0035] Step S430: Perform bioregulatory factor conversion processing according to the fermentation inhibition intensity sequence. By converting the inhibition intensity into a digestive system function correction coefficient, a seasonal inhibition factor is obtained.

[0036] Preferably, in this step, the fermentation inhibition intensity sequence is converted into a correction parameter applicable to the reproduction-metabolism coupling model. Through piecewise linear transformation, the theoretical inhibition rate is mapped to a multiplicative correction coefficient in the range of 0.2 - 0.9 (inhibition factor = 1 indicates no inhibition, and 0.2 indicates 80% inhibition). This step establishes a complete link of "vegetation chemical inhibition - digestive function attenuation - model correction coefficient", making the seasonal effect compensation have both biological significance adaptability and computational operability. This design realizes the transformation of the biochemical interaction law between plants and animals in the ecosystem into computable model parameters, greatly improving the robustness of the prediction model in cross-seasonal scenarios.

[0037] Furthermore, step S500 includes steps S510 to S530.

[0038] Step S510: Perform matrix compensation according to the progesterone-metabolism coupling matrix and the seasonal inhibition factor. By performing weighted product operations on each element of the coupling matrix and the inhibition factor, seasonal effect attenuation correction is achieved, and a seasonally corrected coupling matrix is obtained; Specifically, in this step, seasonal interference correction is performed on the progesterone-metabolism coupling matrix. Through weighted product operations, the inhibition factor is incorporated into each element of the matrix to achieve seasonal effect attenuation correction. By establishing a scalar multiplication rule between the inhibition factor and the matrix elements, the interference effect of bamboo secondary substances on the metabolism-hormone association is eliminated. This step converts the vegetation chemical inhibition effect into a linear compensation parameter in the model calculation space, and the output seasonally corrected coupling matrix significantly improves cross-seasonal stability.

[0039] Step S520: Perform dynamic feature extraction according to the seasonally corrected coupling matrix. By calculating the first derivative sequence and standard deviation analysis within a sliding time window to identify continuous jump intervals, and at the same time detecting the falling position intervals of stability indicators, a reproductive feature marker sequence is obtained; It should be noted that in this step, the seasonally corrected coupling matrix is analyzed within a preset time window (adapted to the estrus cycle), and dual-channel feature capture is performed: based on the calculation of the first derivative sequence, continuous jump intervals exceeding the baseline fluctuation are identified (characterizing sharp changes in hormone flux); simultaneously, the cliff-like falling intervals of the system stability indicators are detected (reflecting the state transition of the reproductive system). By defining the thresholds of jump persistence and stability cliff depth, a reproductive feature marker sequence with both time position and feature intensity is generated.

[0040] Step S530: Perform critical state determination according to the reproductive feature marker sequence. Through a dual-feature spatio-temporal overlap detection mechanism, a binary state marker sequence is generated, and an estrus critical state eigenvalue is obtained.

[0041] Specifically, this step performs a dual-characteristic spatio-temporal coupling analysis on the reproductive characteristic marker sequence: when the jump duration completely contains the stability cliff time point, the binary state marker is activated; otherwise, it is set to zero. Preferably, by constructing a spatio-temporal overlap discriminator, an estrus critical state eigenvalue containing only 0 / 1 is output. This design accurately captures the co-triggering event of the hormone pulse and the system phase transition, achieving high-precision event marking.

[0042] Furthermore, step S600 includes step S610 to step S630.

[0043] Step S610: Perform multi-source feature integration processing based on the bamboo fiber metabolic efficiency sequence, progesterone-metabolism coupling matrix, seasonal inhibition factor, and estrus critical state eigenvalue. Unify the heterogeneous data to the same time reference through a synchronization mechanism that eliminates time mismatch, obtaining a time-normalized feature set. In this step, four types of heterogeneous data, namely the bamboo fiber metabolic efficiency sequence, progesterone-metabolism coupling matrix, inhibition factor, and critical eigenvalue, are calibrated in the time domain through a time baseline unification mechanism. The core technology is to use resampling interpolation methods to eliminate differences in data acquisition frequencies and construct a feature cube (time point × feature type × feature depth) with strict synchronization in the time dimension. This operation solves the problem of modeling distortion caused by "time mismatch" in traditional multi-source physiological data, forming a standardized time-normalized feature set and laying the foundation for collaborative analysis.

[0044] Step S620: Perform dynamic weight configuration processing based on the time-normalized feature set. Construct collaborative prediction features through the correlation relationship between the metabolic basic signal, hormone response intensity, biological inhibition factor, and critical signal, obtaining a weighted prediction feature vector. It can be understood that based on the time-normalized feature set, this step establishes a differential feature contribution degree allocation mechanism: assign a core weight to the metabolic efficiency (reflecting the basic state of the digestive system), a medium weight to the hormone response (characterizing the dynamics of reproductive regulation), a correction weight to the inhibition factor (eliminating seasonal interference), and a trigger weight to the critical signal (event-driven factor). Quantify the weight allocation through principal component contribution rate analysis, generating a weighted prediction feature vector that fuses four-dimensional features. Convert the stacking of static features into a dynamic collaborative architecture that conforms to physiological logic, enabling the metabolic basis and reproductive events to achieve weight self-adaptive matching.

[0045] Step S630: Perform window period probability sequence modeling processing based on the weighted prediction feature vector. Extract the estrus state distribution model in the continuous time dimension through the periodic state evolution law and the probability mapping mechanism, obtaining a dynamic prediction model for the estrus window of the giant panda.

[0046] It should be noted that this step takes the weighted feature vector as input and extracts the periodic state transition law through a sliding time window (window width: 7 days): for the features within the window, the estrus probability density is dynamically calculated, mapped to the probability value in the interval [0, 1] through the Sigmoid function, and the estrus state probability distribution curve for the next 168 hours is continuously output. The model automatically fuses new data every 24 hours for rolling update, generating a dynamic prediction map of the estrus window with both time continuity and confidence (horizontal axis = future time, vertical axis = estrus probability), achieving a qualitative breakthrough from discrete physiological parameters to continuous reproductive event prediction.

[0047] Specifically, as Figure 4 shown, first, the model adopts a dual-channel hybrid architecture. The temporal convolutional network path (TCN path) uses eight layers of dilated causal convolution, and its dilation coefficient is designed to increase in powers of two. Through the residual connection structure, the vanishing gradient of the deep network is avoided, and it is specifically used to capture the long-range trend features with a cycle of about seven days, such as the metabolic efficiency of bamboo fiber and inhibitory factors. The gated recurrent unit path (GRU path) is set with a three-layer bidirectional structure to learn the short-term temporal dependence of progesterone response and critical signals (adapting to the propagation law of the hormone lag effect within 72 hours). The features output by the two paths are concatenated and then reduced in dimension through a fully connected layer to form a one-dimensional probability value output. This architecture introduces an anatomical and physiological adaptation mechanism: the depth setting of the temporal convolutional network is precisely matched to the cellulose metabolism cycle of giant pandas, and the bidirectional flow characteristics of the gated recurrent unit effectively simulate the bidirectional regulation process of progesterone's "promoting metabolism - post-estrus inhibition", and the robustness to sudden disturbances in the captive environment is enhanced through the residual structure and hierarchical dropout technology.

[0048] The training data is constructed with the weighted feature vector as input, centered on the true value of the ovulation time, and probability labels are generated using a Gaussian distribution with a standard deviation of 12 hours to simulate the unique "slow climb before ovulation - sharp drop after ovulation" biological pattern of the estrus window. The loss function combines three components: the mean squared error term as the basic prediction error metric, the KL divergence term to constrain the output distribution to conform to the reproductive physiological law, and the peak penalty term to specifically suppress the interference of pseudo-peaks. The training adopts the cosine annealing learning rate strategy. When the validation loss decreases by less than one-thousandth for ten consecutive training cycles, the early stopping mechanism is triggered, and the transfer learning paradigm (pre-training the historical data of the Chengdu base first and then fine-tuning the target individual) is introduced to effectively solve the small sample problem.

[0049] The Gaussian distribution label generation formula is: , where, P p(t) represents the estrus occurrence probability at time t; t represents the time point for which the probability is to be calculated; T represents the true ovulation time; e e represents the natural base; σ represents the standard deviation, taking 12 hours.

[0050] The loss function formula is as follows: ; where L represents the loss function; α represents the weight of the mean square error term; MSE represents the mean square error between the predicted value and the label; β represents the weight of the KL divergence term; KL represents the divergence between the predicted distribution and the biological prior; γ represents the weight of the peak penalty term; PeakPenalty represents the pseudo-peak suppression loss term.

[0051] Finally, the verification system establishes three core indicators: peak time error, curve steepness (maximum value of the probability change rate), and confidence level (area under the curve three hours before and after the ovulation moment). The clinical deployment requirements are no greater than 6 hours, no less than 0.15 per hour, and no less than 0.9 respectively. The deployment mechanism is updated every 24 hours in a rolling manner: receiving new physiological data - performing feature updates - sliding window prediction for the next 168 hours. The output decision includes three elements: the best operation time window, real-time confidence level, and alert level, and it can complete a single prediction within three seconds through lightweight compression, providing a practical edge computing solution for the wildlife reserve. Adversarial training injects ±20% fecal weight noise samples to improve robustness, and an attention gating mechanism is designed (the gating signal is the sigmoid function of the concatenated vector of the weight matrix, the hidden state of the gated recurrent unit, and the critical feature). When the critical feature is activated, the hormone path weight is automatically increased by 37%. The attention gating formula is expressed as: where, Attention represents the dynamic enhancement gate; W c represents the learnable weight matrix; h GRU represents the hidden state vector of the gated recurrent unit path; F critical represents the critical feature activation vector; σ1(·) represents the sigmoid activation function.

[0052] Example 2:

[0053] As Figure 2 shown, this example provides a system for modeling the relationship between fecal weight change and the estrous cycle of giant pandas. The system includes: An acquisition module 901, configured to acquire the basic monitoring data of female giant panda individuals, including daily fecal weight measurement values, serum progesterone concentration detection values, and natural season classification labels; An extraction module 902, configured to extract fiber metabolism characteristics based on the daily fecal weight measurement values to obtain a metabolic efficiency sequence of bamboo fibers; A construction module 903, configured to construct a progesterone-metabolism coupling matrix based on the metabolic efficiency sequence and the serum progesterone concentration detection values; A correction module 904, configured to perform correction based on the natural season classification labels, and obtain an inhibition factor by quantifying the inhibitory effect of the seasonal fluctuation of bamboo leaf secondary metabolites on the fermentation rate; An identification module 905, configured to identify based on a progesterone-metabolism coupling matrix and an inhibitory factor, and obtain an estrus critical state eigenvalue by detecting the derivative jump characteristics and stability characteristics of the corrected coupling matrix within a preset time window; An output module 906, configured to construct a dynamic prediction model for the estrus window of giant pandas based on a metabolic efficiency sequence, a progesterone-metabolism coupling matrix, an inhibitory factor, and an estrus critical state eigenvalue.

[0054] In a specific embodiment disclosed in the present application, the extraction module 902 includes: A first extraction unit, configured to separate feeding fluctuations according to daily fecal weight measurement values, and separate abnormal feeding frequency fluctuation components through moving standard deviation time domain decomposition to obtain a basal metabolism component sequence; A second extraction unit, configured to extract cecal rhythm according to the basal metabolism component sequence, and extract the energy value of the characteristic frequency band of cecal cellulose fermentation through wavelet packet analysis to obtain a cecal rhythm energy sequence; A third extraction unit, configured to perform metabolic efficiency integration processing according to the basal metabolism component sequence and the cecal rhythm energy sequence, and obtain a bamboo fiber metabolic efficiency sequence by analyzing the synergistic mechanism between cecal movement and steady-state metabolism and correcting the influence of biological sample physical property variation.

[0055] In a specific embodiment disclosed in the present application, the construction module 903 includes: A first construction unit, configured to quantify the metabolic fluctuation pattern of the digestive tract metabolism according to the bamboo fiber metabolic efficiency sequence, and obtain a reproductive cycle metabolic characteristic sequence by extracting the periodic characteristics of metabolic efficiency and the precursor signal of the critical point; A second construction unit, configured to evaluate the hormone time-delay effect according to the detected value of serum progesterone concentration, and obtain a progesterone regulation kinetics sequence by identifying the delayed regulation law of progesterone concentration gradient change on intestinal function; A third construction unit, configured to perform biological coupling relationship modeling according to the reproductive cycle metabolic characteristic sequence and the progesterone regulation kinetics sequence, and obtain a progesterone-metabolism coupling matrix by constructing a hormone-metabolism dose response surface to characterize three-dimensional association.

[0056] In a specific embodiment disclosed in the present application, the correction module 904 includes: A first correction unit, configured to perform dynamic analysis of bamboo secondary metabolites according to natural season classification labels, and obtain a secondary metabolite concentration characteristic sequence by establishing a mapping relationship between the seasonal cycle and the fluctuation of leaf tannic acid concentration; A second correction unit, configured to perform digestion inhibition effect evaluation processing according to the secondary metabolite concentration characteristic sequence, and obtain a fermentation inhibition intensity sequence by analyzing the dose inhibition effect of tannin substances on the cellulase activity of cecal microorganisms. A third correction unit, configured to perform bioregulatory factor conversion processing according to the fermentation inhibition intensity sequence, and obtain seasonal inhibition factors by converting the inhibition intensity into a digestive system function correction coefficient.

[0057] In a specific embodiment disclosed in the present application, the recognition module 905 includes: A first recognition unit, configured to perform matrix compensation according to the progesterone-metabolism coupling matrix and the seasonal inhibition factors, and realize seasonal effect attenuation correction by performing weighted product operations on the elements of the coupling matrix and the inhibition factors, so as to obtain a seasonally corrected coupling matrix; A second recognition unit, configured to perform dynamic feature extraction according to the seasonally corrected coupling matrix, identify a persistent jump interval by performing first derivative sequence calculation and standard deviation analysis within a sliding time window, and simultaneously detect the falling position interval of the stability index, so as to obtain a reproductive feature marker sequence; A third recognition unit, configured to perform critical state determination according to the reproductive feature marker sequence, generate a binary state marker sequence through a dual feature spatio-temporal overlap detection mechanism, and obtain an estrus critical state eigenvalue.

[0058] In a specific embodiment disclosed in the present application, the output module 906 includes: A first output unit, configured to perform multi-source feature integration processing according to the bamboo fiber metabolism efficiency sequence, the progesterone-metabolism coupling matrix, the seasonal inhibition factors, and the estrus critical state eigenvalue, and unify heterologous data to the same time reference through a synchronization mechanism for eliminating time mismatch, so as to obtain a time-reference-based feature set; A second output unit, configured to perform dynamic weight configuration processing according to the time-reference-based feature set, and construct a collaborative prediction feature through the correlation relationship between the metabolic basic signal, the hormone response intensity, the biological inhibition factor, and the critical signal, so as to obtain a weighted prediction feature vector; A third output unit, configured to perform window period probability sequence modeling processing according to the weighted prediction feature vector, and form an estrus state distribution model in a continuous time dimension through a periodic state evolution rule extraction and probability mapping mechanism, so as to obtain a dynamic prediction model for the estrus window of giant pandas.

[0059] Embodiment 3:

[0060] Corresponding to the above method embodiment, in this embodiment, a device for associating the change in fecal weight with the estrus cycle of giant pandas is also disclosed. A device for associating the change in fecal weight with the estrus cycle of giant pandas described below can be mutually referred to with a method for associating the change in fecal weight with the estrus cycle of giant pandas described above.

[0061] Figure 3is a block diagram of a modeling device 800 for associating fecal weight changes with the estrus cycle of giant pandas shown according to an exemplary embodiment. As Figure 3 shown, the modeling device 800 for associating fecal weight changes with the estrus cycle of giant pandas may include: a processor 801, a memory 802. The modeling device 800 for associating fecal weight changes with the estrus cycle of giant pandas may further include one or more of a multimedia component 803, an I / O interface 804, and a communication component 805.

[0062] Among them, the processor 801 is used to control the overall operation of the modeling device 800 for associating fecal weight changes with the estrus cycle of giant pandas, so as to complete all or part of the steps in the above-mentioned method for associating fecal weight changes with the estrus cycle of giant pandas. The memory 802 is used to store various types of data to support the operation of the modeling device 800 for associating fecal weight changes with the estrus cycle of giant pandas. These data may include, for example, instructions for any application or method operating on the modeling device 800 for associating fecal weight changes with the estrus cycle of giant pandas, as well as application-related data, such as contact data, sent and received messages, pictures, audio, video, and so on. The memory 802 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic memory, flash memory, a magnetic disk, or an optical disc. The multimedia component 803 may include a screen and an audio component. The screen may be a touch screen, for example, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone, and the microphone is used to receive external audio signals. The received audio signals may be further stored in the memory 802 or sent through the communication component 805. The audio component further includes at least one speaker for outputting audio signals. The I / O interface 804 provides an interface between the processor 801 and other interface modules, and the above-mentioned other interface modules may be a keyboard, a mouse, buttons, etc. These buttons may be virtual buttons or physical buttons. The communication component 805 is used for the modeling device 800 for associating fecal weight changes with the estrus cycle of giant pandas to communicate with other devices in a wired or wireless manner. Wireless communication, such as Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G, or 4G, or a combination of one or more of them. Accordingly, the communication component 805 may include: a Wi-Fi module, a Bluetooth module, and an NFC module.

[0063] In an exemplary embodiment, a modeling device 800 for associating fecal weight changes with the estrus cycle of giant pandas may be implemented by one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components, and is used to execute the above-mentioned method for associating fecal weight changes with the estrus cycle of giant pandas.

[0064] In another exemplary embodiment, a computer-readable storage medium including program instructions is further provided. When the program instructions are executed by a processor, the steps of the above-mentioned method for associating fecal weight changes with the estrus cycle of giant pandas are implemented. For example, the computer-readable storage medium may be the above-mentioned memory 802 including program instructions, and the above program instructions may be executed by the processor 801 of a modeling device 800 for associating fecal weight changes with the estrus cycle of giant pandas to complete the above-mentioned method for associating fecal weight changes with the estrus cycle of giant pandas.

[0065] As described above, the above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered by the protection scope of the present invention.

Claims

1. A method for establishing a correlation model between fecal weight changes and the estrus cycle of giant pandas, characterized in that Including: Obtaining basic monitoring data of female giant panda individuals, including daily fecal weight values, serum progesterone concentration detection values, and natural season classification labels; Extracting fiber metabolism characteristics based on the daily fecal weight values to obtain a metabolic efficiency sequence of bamboo fibers; Constructing a matrix based on the metabolic efficiency sequence and the serum progesterone concentration detection values to obtain a progesterone-metabolism coupling matrix; Making corrections according to the natural season classification labels, and obtaining an inhibition factor by quantifying the inhibitory effect of seasonal fluctuations of bamboo leaf secondary metabolites on the fermentation rate; Identifying based on the progesterone-metabolism coupling matrix and the inhibition factor, and obtaining an estrus critical state eigenvalue by detecting the derivative jump characteristics and stability characteristics of the corrected coupling matrix within a preset time window; Constructing a model based on the metabolic efficiency sequence, the progesterone-metabolism coupling matrix, the inhibition factor, and the estrus critical state eigenvalue to obtain a dynamic prediction model for the estrus window of giant pandas.

2. The method for establishing a model related to the association between fecal weight change and the estrus cycle of giant pandas according to claim 1, wherein Extracting fiber metabolism characteristics based on the daily fecal weight values to obtain a metabolic efficiency sequence of bamboo fibers, including: Separating feeding fluctuations based on the daily fecal weight values, and separating abnormal feeding frequency fluctuation components through moving standard deviation time domain decomposition to obtain a basic metabolism component sequence; Extracting cecal rhythm based on the basic metabolism component sequence, and extracting the energy value of the cecal cellulose fermentation characteristic frequency band through wavelet packet analysis to obtain a cecal rhythm energy sequence; Performing metabolic efficiency integration processing on the basic metabolism component sequence and the cecal rhythm energy sequence, and obtaining a bamboo fiber metabolic efficiency sequence by analyzing the synergistic mechanism of cecal movement and steady-state metabolism and correcting the influence of biological sample physical property variation.

3. The method for establishing a model related to the association between fecal weight change and the estrus cycle of giant pandas according to claim 1, wherein Constructing a matrix based on the metabolic efficiency sequence and the serum progesterone concentration detection values to obtain a progesterone-metabolism coupling matrix, including: Quantifying the metabolic fluctuation pattern of the digestive tract metabolism state based on the bamboo fiber metabolic efficiency sequence, and obtaining a reproductive cycle metabolic characteristic sequence by extracting the periodic characteristics of metabolic efficiency and the precursor signal of the critical point; Evaluating the hormone time-delay effect based on the serum progesterone concentration detection values, and obtaining a progesterone regulation kinetic sequence by identifying the delayed regulation law of progesterone concentration gradient change on intestinal function; Modeling the biological coupling relationship between the reproductive cycle metabolic characteristic sequence and the progesterone regulation kinetic sequence, and obtaining a progesterone-metabolism coupling matrix by constructing a hormone-metabolism dose response surface to characterize the three-dimensional association.

4. The method for establishing a model associating fecal weight change with the estrus cycle of giant pandas according to claim 1, wherein Making corrections according to the natural season classification labels, and obtaining an inhibition factor by quantifying the inhibitory effect of seasonal fluctuations of bamboo leaf secondary metabolites on the fermentation rate, including: Performing dynamic analysis of bamboo secondary substances according to the natural season classification labels, and obtaining a secondary metabolite concentration characteristic sequence by establishing a mapping relationship between the seasonal cycle and the fluctuation of leaf tannic acid concentration; Evaluating the digestive inhibition effect based on the secondary metabolite concentration characteristic sequence, and obtaining a fermentation inhibition intensity sequence by analyzing the dose inhibition effect of tannin substances on the cellulase activity of cecal microorganisms; Perform bioregulatory factor conversion processing according to the fermentation inhibition intensity sequence. By converting the inhibition intensity into a digestive system function correction coefficient, a seasonal inhibition factor is obtained.

5. The method for establishing a model related to the association between fecal weight change and the estrus cycle of giant pandas according to claim 1, wherein Perform identification according to the progesterone-metabolism coupling matrix and the inhibition factor. By detecting the derivative jump characteristics and stability characteristics of the corrected coupling matrix within a preset time window, an estrus critical state eigenvalue is obtained, including: Perform matrix compensation according to the progesterone-metabolism coupling matrix and the seasonal inhibition factor. By performing weighted product operations on each element of the coupling matrix and the inhibition factor to achieve seasonal effect attenuation correction, a seasonally corrected coupling matrix is obtained; Perform dynamic feature extraction according to the seasonally corrected coupling matrix. By calculating the first derivative sequence and performing standard deviation analysis within a sliding time window to identify a persistent jump interval, and at the same time detecting the falling position interval of the stability index, a reproductive feature marker sequence is obtained; Perform critical state determination according to the reproductive feature marker sequence. By generating a binary state marker sequence through a dual feature spatio-temporal overlap detection mechanism, an estrus critical state eigenvalue is obtained.

6. A modeling system for associating fecal weight changes with the estrous cycle of giant pandas, characterized in that, Including: An acquisition module, used to acquire the basic monitoring data of female giant panda individuals, including daily fecal weight measurement values, serum progesterone concentration detection values, and natural season classification labels; An extraction module, used to extract fiber metabolism characteristics according to the daily fecal weight measurement values to obtain a metabolic efficiency sequence of bamboo fiber; A construction module, used to construct a matrix according to the metabolic efficiency sequence and the serum progesterone concentration detection values to obtain a progesterone-metabolism coupling matrix; A correction module, used to perform correction according to the natural season classification label. By quantifying the inhibitory effect of seasonal fluctuations of bamboo leaf secondary metabolites on the fermentation rate, an inhibition factor is obtained; An identification module, used to perform identification according to the progesterone-metabolism coupling matrix and the inhibition factor. By detecting the derivative jump characteristics and stability characteristics of the corrected coupling matrix within a preset time window, an estrus critical state eigenvalue is obtained; An output module, used to construct a model according to the metabolic efficiency sequence, the progesterone-metabolism coupling matrix, the inhibition factor, and the estrus critical state eigenvalue to obtain a dynamic prediction model for the estrus window of giant pandas.

7. The fecal weight change and giant panda estrus cycle associated modeling system according to claim 6, characterized in that The extraction module includes: A first extraction unit, used to separate feeding fluctuations according to the daily fecal weight measurement values. By performing time domain decomposition of the moving standard deviation to separate abnormal feeding frequency fluctuation components, a basic metabolism component sequence is obtained; A second extraction unit, used to extract cecal rhythm according to the basic metabolism component sequence. By performing wavelet packet analysis to extract the energy value of the cecal cellulose fermentation characteristic frequency band, a cecal rhythm energy sequence is obtained; A third extraction unit, used to perform metabolic efficiency integration processing according to the basic metabolism component sequence and the cecal rhythm energy sequence. By analyzing the synergistic mechanism of cecal movement and steady-state metabolism and correcting the influence of biological sample physical property variation, a metabolic efficiency sequence of bamboo fiber is obtained.

8. The fecal weight change and giant panda estrus cycle associated modeling system according to claim 6, wherein The construction module includes: The first construction unit is used to quantify the metabolic fluctuation pattern of the digestive tract metabolism state according to the bamboo fiber metabolic efficiency sequence, and obtain the reproductive cycle metabolic characteristic sequence by extracting the periodic characteristics of metabolic efficiency and the precursor signal of the critical point. The second construction unit is used to evaluate the hormone time-delay effect according to the detected value of serum progesterone concentration, and obtain the progesterone regulation kinetic sequence by identifying the delayed regulation law of the progesterone concentration gradient change on intestinal function. The third construction unit is used to model the biological coupling relationship according to the reproductive cycle metabolic characteristic sequence and the progesterone regulation kinetic sequence, and obtain the progesterone-metabolism coupling matrix by constructing a hormone-metabolism dose-response surface to characterize the three-dimensional association.

9. The fecal weight change and giant panda estrus cycle associated modeling system according to claim 6, wherein The correction module includes: The first correction unit is used to dynamically analyze the secondary substances of bamboo plants according to the natural season classification label, and obtain the secondary metabolite concentration characteristic sequence by establishing the mapping relationship between the season cycle and the fluctuation of leaf tannic acid concentration. The second correction unit is used to evaluate the digestive inhibition effect according to the secondary metabolite concentration characteristic sequence, and obtain the fermentation inhibition intensity sequence by analyzing the dose inhibition effect of tannin substances on the cellulase activity of cecal microorganisms. The third correction unit is used to perform biological regulator conversion processing according to the fermentation inhibition intensity sequence, and obtain the seasonal inhibition factor by converting the inhibition intensity into the digestive system function correction coefficient.

10. The fecal weight change and giant panda estrus cycle associated modeling system according to claim 6, wherein The recognition module includes: The first recognition unit is used to perform matrix compensation according to the progesterone-metabolism coupling matrix and the seasonal inhibition factor, and obtain the season-corrected coupling matrix by performing weighted product operation on each element of the coupling matrix and the inhibition factor to achieve attenuation correction of the season effect. The second recognition unit is used to extract dynamic features according to the season-corrected coupling matrix, and obtain the reproductive characteristic marker sequence by calculating the first derivative sequence and performing standard deviation analysis within the sliding time window to identify the continuous jump interval, and simultaneously detecting the falling position interval of the stability index. The third recognition unit is used to determine the critical state according to the reproductive characteristic marker sequence, and obtain the estrus critical state characteristic value by generating a binary state marker sequence through a dual-characteristic spatio-temporal overlap detection mechanism.

Images