Portable animal body component analyzer

Abstract

The invention relates to the technical field of animal body component detection, and provides a portable animal body component analyzer, which comprises a data acquisition module, a data processing module, an analysis module, a data processing module, an analysis module and an analysis module, wherein the data acquisition module is responsible for acquiring multi-frequency bioelectrical impedance data and near infrared spectrum data of domesticated pets; the data processing module is used for carrying out noise filtering, dynamic baseline correction and normalization processing on the collected original data; the algorithm analysis module uses a variety inference and dynamic correction method to correct data errors, extracts body fat distribution, tissue metabolism and body fluid distribution characteristics through a multi-modal data fusion method, and calculates growth and development health scores through an integrated learning supervision model; the database management module stores individual information, appearance characteristics, measurement data and analysis results, and supports data query and backtracking; the user interface is used for inputting animal information and hair states and displaying analysis results, variety similarity, body composition change trends and health scores.

Description

Technical Field

[0001] The present invention relates to the technical field of animal body composition detection, and more specifically, the present invention relates to a portable animal body composition analyzer. Background Art

[0002] Body composition analysis is a key technology for evaluating the fat, lean body mass (mainly composed of protein and water), and bone mineral content of organisms. By measuring these parameters, the individual's health status, growth and development level, and metabolic function can be deeply understood. For domestic pets in the growth and development stage, the dynamic changes in body composition can directly reflect their growth and development status and nutritional health level. Domestic pets experience obvious changes in body composition during the growth and development stage, including fat accumulation, protein synthesis, and the regulation of body fluid balance. Therefore, accurate measurement of their body composition is crucial for understanding the growth and development process and formulating a scientific feeding plan.

[0003] Literature 1 (Scholz AM, et al, Non-invasive methods for the determination of body and carcass composition in livestock. Animal, 2015) points out that body composition analysis technology plays an important role in livestock breeding optimization, health assessment, and scientific research. This study details non-invasive techniques including ultrasound (US), dual-energy X-ray absorptiometry (DXA), computed tomography (CT), and magnetic resonance imaging (MRI). These techniques can measure the distribution and volume of tissue fat, lean meat, and bone with high precision, providing important support for optimizing animal health management and improving production performance. In particular, the DXA technology is widely recognized for its portability and efficiency, but its equipment complexity and high cost limit its use in daily monitoring.

[0004] Reference 2 (Karelis AD, et al, Validation of a portable bioelectrical impedance analyzer for the assessment of body composition. Applied Physiology, Nutrition, and Metabolism, 2013) studied the accuracy and applicability of a portable bioelectrical impedance analyzer (Inbody 230). The research results showed that through multi-frequency impedance measurement technology, this device could efficiently assess human fat mass, body fat percentage, and lean body mass, and its measurement results had a high correlation with DXA (r = 0.94–0.99). Nevertheless, there were systematic biases in the assessment of trunk and limb lean body mass by this device. Portable devices are particularly suitable for daily monitoring scenarios outside the laboratory due to their ease of operation and high precision.

[0005] Although existing body composition analysis techniques, such as DXA and portable bioelectrical impedance analyzers, can meet certain research needs, they still have significant limitations when faced with the special requirements of dynamically monitoring the body composition of domestic pets during their growth and development periods. For example, existing devices have insufficient ability to fuse and analyze multi-modal data and cannot fully integrate information on the distribution of fat, protein, and body fluids. These technical deficiencies limit the comprehensive understanding and accurate monitoring of the dynamic changes in the body composition of domestic pets during their growth and development periods. Moreover, due to significant differences in body size, hair characteristics, and body composition distribution among different breeds of domestic pets, the lack of a correction mechanism based on breed characteristics limits the accuracy of measurement results. In particular, existing devices do not adequately consider the influence of the hair state of pets and cannot achieve effective compensation for the hairy state and shaved state, resulting in measurement deviations. At the same time, the lack of the ability to infer the body composition of pets of unknown breeds makes it difficult to provide accurate assessments for crossbred animals. Summary of the Invention

[0006] To overcome the above-mentioned defects of the prior art, the present invention provides a portable animal body composition analyzer. Through multi-frequency bioelectrical impedance and near-infrared spectroscopy data collection, using a breed inference unit and a dynamic correction unit for time series prediction and compensation, combining a multi-modal data fusion method to extract body fat distribution, tissue metabolism, and body fluid distribution characteristics, and calculating a growth and development health score based on an integrated learning supervision model, it solves the problems of the prior art in pet breed-specific adaptation, compensation for the influence of hair state, dynamic correction during the growth and development period, and insufficient multi-modal data fusion analysis ability, and realizes the accurate monitoring and health assessment of the body composition of domestic pets during their growth and development periods.

[0007] To achieve the above object, the present invention provides the following technical solutions: A portable animal body composition analyzer includes a data acquisition module, a data processing module, a database management module, an algorithm analysis module, and a user interface.

[0008] As a further solution of the present invention, the data acquisition module is connected to the data processing module, the data processing module is connected to the algorithm analysis module, the algorithm analysis module is bidirectionally connected to the database management module, the algorithm analysis module is connected to the user interface, and the database management module is connected to the user interface; The data acquisition module is used to collect the original data of the body composition of domestic pets from a bioelectrical impedance device and a near-infrared spectroscopy device, and the bioelectrical impedance device and the near-infrared spectroscopy device are configured for portable devices; bioelectrical impedance technology is used to measure the impedance of body tissues to weak currents, and near-infrared spectroscopy technology is used to analyze the reflection or absorption characteristics of light by different tissues.

[0009] The algorithm analysis module is used to analyze and calculate the preprocessed bioelectrical impedance data and near-infrared spectroscopy data, including a breed inference unit, a dynamic correction unit, a body composition analysis unit, and a health scoring unit, where: the breed inference unit is based on appearance feature data, and combines a known breed database or cluster analysis to analyze pets of known and unknown breeds, and generates breed similarity information and a breed inference adjustment factor; The dynamic correction unit is used to perform dynamic correction on the body composition data of domestic pets by using a time series model based on historical data and combining the breed inference adjustment factor, and is also used to calculate a hair state compensation coefficient according to the hair state information; The dynamic correction unit is configured to perform dynamic correction on the body composition data during the growth and development of domestic pets by using a time series model based on historical data and combining the breed inference adjustment factor, and output the corrected body composition data; the dynamic correction unit is also configured to receive the hair state information input by the user, calculate the hair state compensation coefficient according to the hair state information and the breed inference adjustment factor, and apply the hair state compensation coefficient to the correction process of the body composition measurement data, where the hair state information includes a hair state identifier and hair characteristic parameters; The body composition analysis unit is configured to receive the corrected data and analyze the body fat distribution, tissue metabolism, and body fluid distribution indexes of domestic pets through a multimodal data fusion method; the health scoring unit is configured to evaluate the health status of domestic pets during the growth and development period based on an integrated learning supervision model, and calculate the growth and development health score of an individual, where the score calculation is adjusted by weight in combination with the output result of the breed inference unit.

[0010] As a further solution of the present invention, the measurement points of the data acquisition module are divided into core measurement points and auxiliary measurement points. The core measurement points include the midline of the neck, the midline of the chest, the midline of the abdomen, the midline of the back, and characteristic measurement points determined according to the animal species. During measurement, positioning is based on anatomical landmark points; the auxiliary measurement points include the two side points of the back and the distal points of the limbs, and their positions are determined according to the relative positions of the core measurement points; after the data acquisition module obtains the core measurement point data, preliminary analysis can be completed, and the auxiliary measurement point data is collected when higher accuracy is required.

[0011] As a further solution of the present invention, the breed inference unit is configured to perform the following operations: Receive the pet appearance feature data input by the user through the user interface. The appearance features include body size, body proportion, hair length, hair color, ear shape, and tail characteristics; Convert the appearance feature data into a feature vector, compare it with the known breed feature vector library in the database, and calculate the similarity; Determine the processing path according to the similarity. When the similarity is higher than the preset threshold, use the random forest classification algorithm for breed classification; when the similarity is lower than the preset threshold, use the K-Means clustering algorithm for body type category division; the training data of the random forest classification algorithm comes from the appearance feature vectors in the known breed database; Calculate the breed inference adjustment factor according to the result of breed classification or body type category division for the calculation of the subsequent processing unit.

[0012] As a further solution of the present invention, the breed adjustment factor is a weight coefficient calculated based on breed similarity. The generation steps of the breed inference adjustment factor include: Step 1: Extract the feature vector according to the pet appearance features input by the user; Step 2: Calculate the cosine similarity between the feature vector and the feature vectors of each breed in the known breed database; Step 3: Take the breed corresponding to the maximum similarity as the candidate breed and calculate the adjustment factor, where: ; In the formula, is the maximum similarity, is the adjustment factor.

[0013] As a further solution of the present invention, the dynamic correction unit is configured to perform the following operations: Establish a historical database to store the body composition measurement data of domestic pets at different growth and development stages. The measurement data includes bioelectrical impedance data and near-infrared spectroscopy data reflecting body surface fat content, protein content, and water content; Perform time series analysis on the historical data, extract the characteristic change patterns of each growth and development stage, and construct an error compensation model; Input the data of each measurement point collected into the error compensation model to generate the expected parameter values of body fat content, protein content, and moisture content; Compare the expected parameter values of body fat content, protein content, and moisture content with the actual measured values, and calculate the error values of each index; Construct a compensation function for correction based on the error value and the breed inference adjustment factor output by the breed inference unit, and the compensation function takes into account the relationship between the error vector, the breed standard weight, and the actual weight; Apply the compensation function to the real-time data of the measurement point to generate the corrected numerical values of each index, and store the correction result in the historical database.

[0014] The time series model uses an improved LSTM network. Its input layer includes historical body composition data (body fat rate, protein content, moisture content) and the breed inference adjustment factor. The hidden layer is set with a bidirectional gated recurrent unit (BiGRU), and the output layer generates an error compensation value through a fully connected network. When training the model, the sliding window method is used to segment the historical data. The window length is 3 months, the step size is 7 days, and the loss function is a weighted combination of MAE and the breed weight.

[0015] As a further solution of the present invention, the dynamic correction unit is further configured to perform the following operations: Receive the hair status identifier input by the user through the user interface, and the hair status identifier is used to indicate whether the pet is in a shaved state or a hairy state currently; When the hair status identifier is in the hairy state, retrieve the data of the hair compensation coefficient library from the database management module; According to the breed inference adjustment factor output by the breed inference unit and the hair length grade and hair density characteristic value in the appearance feature data, calculate the hair status compensation coefficient and the dynamic compensation value; the corrected body composition data Is corrected to: ; In the formula, Is the original measured value, Is the hair status compensation coefficient, Is the dynamic compensation value, Is the breed inference adjustment factor; Output the corrected body composition data to the body composition analysis unit for subsequent analysis.

[0016] Method for constructing the hair compensation coefficient library: Collect the body composition data of the same pet in the shaved and hairy states to form a paired sample set; Calculate the hair interference compensation equation through a linear regression model: ; wherein, is the hair length grade (Level 1 - 5), is the eigenvalue of hair density; Pre - store the compensation coefficients of different breeds into the database, and the index key is breed ID + hair status.

[0017] As a further solution of the present invention, the body composition analysis unit is configured to perform the following operations: Receive the corrected data and the original data output by the dynamic correction unit; Extract the impedance characteristic parameters of each measurement point from the bioelectrical impedance data and construct a body fat distribution feature vector; Extract the spectral features related to proteins from the near - infrared spectral data and construct a tissue metabolism feature vector; Extract the features related to moisture from the near - infrared spectral data and construct a body fluid distribution feature vector; Perform multi - modal data fusion on the three obtained feature vectors by using a fusion method of weighted average or feature splicing, and the fusion process combines each feature vector through weight coefficients; Calculate body fat percentage index, body fluid distribution index and tissue metabolism index based on the fusion features and input them into the subsequent unit.

[0018] As a further solution of the present invention, the health score unit is configured to perform the following operations: Construct a health assessment index system for the growth and development period of domestic pets, including body fat percentage index, body fluid distribution index and tissue metabolism index; Based on historical data, construct an association model between assessment indexes, and take the output results of the animal category, breed classification and breed inference unit as influencing factors into the model; Use the XGBoost integrated learning method to train the health score model; Input the real - time monitoring data into the health score model and calculate the growth and development health score.

[0019] The hyper - parameters of the XGBoost model are set as follows: learning rate 0.05, tree depth 8, minimum number of samples in a leaf 10, regularization term λ = 1.0, and the input features include body fat percentage index, body fluid distribution index, tissue metabolism index and breed inference adjustment factor.

[0020] As a further aspect of the present invention, the data processing module is configured to perform noise filtering, dynamic baseline correction, and normalization on the collected raw bioelectrical impedance data and near-infrared spectroscopy data. Noise filtering mainly targets high-frequency interference in bioelectrical impedance signals and random noise in spectroscopy data; dynamic baseline correction is used to eliminate baseline drift in near-infrared spectroscopy data; normalization processing makes data from different sources comparable, facilitating subsequent data fusion and analysis.

[0021] As a further aspect of the present invention, the database management module includes: an individual information table for storing basic information and breed information of domestic pets at different growth and development stages; an appearance feature table for storing eigenvectors of standard breeds and appearance feature data of unknown breed pets input by users; a clustering model table, a measurement data table, and an analysis result table; a hair compensation coefficient library for storing hair compensation coefficients under different hair lengths and densities.

[0022] As a further aspect of the present invention, the user interface includes: an animal information input area, including a known breed selection mode and an unknown breed feature input mode. The known breed mode is used to select the animal category, specific breed, and input appearance features, and the unknown breed mode is used to input appearance features such as body size, hair characteristics, ear shape, and tail characteristics; a hair status selection area for selecting the current hair status of the pet, including a shaved state and a hairy state. When the hairy state is selected, the hair length grade and estimated hair density can be further input; a real-time data display area for displaying the values of body fat percentage indicators, body fluid distribution indicators, tissue metabolism indicators, and the breed similarity analysis results of unknown breed pets; a result analysis area for displaying the body composition change trends, standard reference ranges, and growth and development health scores at four growth and development stages.

[0023] Compared with the prior art, the beneficial effects of a portable animal body composition analyzer of the present invention are as follows: By combining the multi-frequency bioelectrical impedance technology and near-infrared spectroscopy technology through the data acquisition module, the present invention can simultaneously obtain multi-modal data of fat, protein, and water of domestic pets, and use the breed inference unit and dynamic correction unit to perform breed-specific adjustment, time series prediction, and error compensation on the measurement data. In contrast, although the bioelectrical impedance analysis devices in the prior art can achieve portable measurement, they only rely on single impedance data for body composition analysis, lacking a compensation mechanism for animal breed differences and hair status, as well as the ability to perform real-time correction for dynamic changes. Especially in the daily use environment, the data accuracy is difficult to guarantee.

[0024] The algorithm analysis module of the present invention adopts a feature fusion algorithm and a Bayesian network model. By integrating multi-modal data to extract body composition distribution features and calculating the growth and development health score, it can directly complete complex data analysis locally without relying on large computing devices or external data support, which further enhances the advantages of the system in terms of portability and independence. Existing technologies such as DXA devices, although having high measurement accuracy, are large in size and high in cost, and are only suitable for use in fixed laboratory environments, making it difficult to meet the needs of daily home monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 FIG. is an example diagram of 3D analysis and reconstruction of lamb image data in Document 1.

[0026] Figure 2 FIG. is a comparison diagram of backfat ultrasound of Iberian pigs and German Large White pigs in Document 1.

[0027] Figure 3 FIG. is a structural display diagram of a flexible dry electrode.

[0028] Figure 4 FIG. is a schematic diagram of the system structure of a portable animal body composition analyzer of the present invention.

[0029] Figure 5 FIG. is a framework diagram of the user interface of a portable animal body composition analyzer of the present invention.

[0030] Figure 6 FIG. is a flowchart of the algorithm in a portable animal body composition analyzer of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0031] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0032] A portable animal body composition analyzer includes a data acquisition module, a data processing module, a database management module, an algorithm analysis module, and a user interface.

[0033] In the embodiments of the present invention, the data acquisition module is connected to the data processing module, the data processing module is connected to the algorithm analysis module, the algorithm analysis module is bidirectionally connected to the database management module, the algorithm analysis module is connected to the user interface, and the database management module is connected to the user interface; The data acquisition module is used to collect the original data of the body composition of domestic pets from a bioelectrical impedance device and a near-infrared spectroscopy device. The bioelectrical impedance device and the near-infrared spectroscopy device are configured for portable devices. The bioelectrical impedance device is used to measure the impedance of body tissues to weak currents, adopts a four-electrode measurement system, and the working frequency range is from 5 kHz to 1 MHz. It can measure impedance values, phase angles, resistance values, and reactance values. The near-infrared spectroscopy device is used to analyze the reflection or absorption characteristics of different tissues to light, the working wavelength range is from 700 nm to 2500 nm, the resolution is 10 nm, and it is measured in a diffuse reflection mode.

[0034] In the embodiment of the present invention, the measurement points of the data acquisition module are divided into core measurement points and auxiliary measurement points. The core measurement points include the midline of the neck (located on the midline of the neck, at the level of cervical vertebrae C4-C5), the center of the chest (located at the midpoint of the sternum), the center of the abdomen (located 2 cm above the umbilicus), the center of the back (located at the junction of the chest and back, about at the level of the T13 vertebra), and characteristic measurement points determined according to the animal species. When measuring, it is positioned based on anatomical landmark points. The auxiliary measurement points include points on both sides of the back and distal points of the limbs, and their positions are determined according to the relative positions of the core measurement points. After the data acquisition module obtains the core measurement point data, preliminary analysis can be completed. When higher accuracy is required, the auxiliary measurement point data is collected.

[0035] In the embodiment of the present invention, the data processing module is configured to perform noise filtering, dynamic baseline correction, and normalization processing on the collected original bioelectrical impedance data and near-infrared spectroscopy data. Noise filtering mainly targets high-frequency interference in bioelectrical impedance signals and random noise in spectral data. Dynamic baseline correction is used to eliminate baseline drift in near-infrared spectroscopy data. Normalization processing makes data from different sources comparable, facilitating subsequent data fusion and analysis.

[0036] The algorithm analysis module is used to analyze and calculate the preprocessed bioelectrical impedance data and near-infrared spectroscopy data, including a breed inference unit, a dynamic correction unit, a body composition analysis unit, and a health score unit.

[0037] The breed inference unit analyzes pets of known breeds and unknown breeds based on appearance feature data and in combination with a known breed database or cluster analysis, and generates breed similarity information and breed inference adjustment factors.

[0038] In the embodiment of the present invention, the breed inference unit is configured to perform the following operations: Receive pet appearance feature data input by the user through the user interface. The appearance features include body size, body proportion, hair length, hair color, ear shape, and tail characteristics. Convert the appearance feature data into feature vectors, compare them with the known breed feature vector library in the database, and calculate the similarity; Determine the processing path according to the similarity. When the similarity is higher than the preset threshold, use the random forest classification algorithm for breed classification; when the similarity is lower than the preset threshold, use the K-Means clustering algorithm for body type classification; the training data of the random forest classification algorithm comes from the appearance feature vectors in the known breed database; According to the result of breed classification or body type classification, calculate the breed inference adjustment factor for the calculation of the subsequent processing unit.

[0039] The breed adjustment factor in the embodiment of the present invention is a weight coefficient calculated based on breed similarity. The generation steps of the breed inference adjustment factor include: Step 1: Extract the feature vector according to the pet appearance features input by the user; Step 2: Calculate the cosine similarity between the feature vector and the feature vectors of each breed in the known breed database; Step 3: Take the breed corresponding to the maximum similarity as the candidate breed, and calculate the adjustment factor α, where: ; In the formula, is the maximum similarity.

[0040] The dynamic correction unit is configured to dynamically correct the body composition data during the growth and development of domestic pets by using a time series model based on historical data and combining the breed inference adjustment factor, and output the corrected body composition data.

[0041] The dynamic correction unit in the embodiment of the present invention is configured to perform the following operations: Establish a historical database to store the body composition measurement data of domestic pets at different growth and development stages. The measurement data includes bioelectrical impedance data and near-infrared spectrum data reflecting body surface fat content, protein content, and water content; Perform time series analysis on the historical data, extract the characteristic change patterns of each growth and development stage, and construct an error compensation model; Input the data of each measurement point collected into the error compensation model to generate the expected body fat content, protein content, and water content parameter values; Compare the expected body fat content, protein content, and water content parameter values with the actual measured values, and calculate the error values of each index; Construct a compensation function for correction based on the error value and the breed inference adjustment factor output by the breed inference unit. The compensation function considers the relationship between the error vector, breed standard weight, and actual weight; Apply the compensation function to the real-time data of the measurement points to generate the corrected values of various indicators, and store the correction results in the historical database.

[0042] The time series model uses an improved LSTM network. Its input layer includes historical body composition data (body fat percentage, protein content, water content) and breed inference adjustment factors. The hidden layer is set with a bidirectional gated recurrent unit (BiGRU), and the output layer generates an error compensation value through a fully connected network. When training the model, the sliding window method is used to segment historical data. The window length is 3 months, the step size is 7 days, and the loss function is a weighted combination of MAE and breed weights.

[0043] The dynamic correction unit is also configured to receive the hair status information input by the user, calculate the hair status compensation coefficient according to the hair status information and the breed inference adjustment factor, and apply the hair status compensation coefficient to the correction process of the body composition measurement data, where the hair status information includes a hair status identifier and hair characteristic parameters.

[0044] The dynamic correction unit in the embodiment of the present invention is also configured to perform the following operations: Receive the hair status identifier input by the user through the user interface, and the hair status identifier is used to indicate whether the pet is in a shaved state or a furry state currently; When the hair status identifier is in the furry state, retrieve the data of the hair compensation coefficient library from the database management module; According to the breed inference adjustment factor output by the breed inference unit and the hair length grade and hair density characteristic value in the appearance feature data, calculate the hair status compensation coefficient and the dynamic compensation value; the corrected body composition data Is corrected to: ; In the formula, Is the original measurement value, Is the hair status compensation coefficient, Is the dynamic compensation value, Is the breed inference adjustment factor; Output the corrected body composition data to the body composition analysis unit for subsequent analysis.

[0045] Method for constructing the hair compensation coefficient library: Collect the body composition data of the same pet in the shaved and furry states to form a paired sample set; Calculate the hair interference compensation equation through a linear regression model: ; Among them, Is the hair length grade (level 1-5), Is the hair density characteristic value; Pre-store the compensation coefficients of different breeds in the database, with the index key being breed ID + hair status.

[0046] Conduct time series analysis on historical data to extract the characteristic change patterns at each growth and development stage: The body composition analysis unit is configured to receive the corrected data and analyze the body fat distribution, tissue metabolism, and body fluid distribution indicators of domestic pets through a multimodal data fusion method.

[0047] As a further solution of the present invention, the body composition analysis unit is configured to perform the following operations: Receive the corrected data and the original data output by the dynamic correction unit; Extract the impedance characteristic parameters of each measurement point from the bioelectrical impedance data and construct a body fat distribution feature vector; Extract the spectral features related to proteins from the near-infrared spectral data and construct a tissue metabolism feature vector; Extract the features related to moisture from the near-infrared spectral data and construct a body fluid distribution feature vector; Perform multimodal data fusion on the three obtained feature vectors using a weighted average or feature splicing fusion method, and the fusion process combines the feature vectors through weight coefficients; Calculate the body fat percentage index, body fluid distribution index, and tissue metabolism index based on the fused features and input them into the subsequent unit.

[0048] The health scoring unit is configured to evaluate the health status of domestic pets during the growth and development period based on an integrated learning supervision model, calculate the growth and development health score of an individual, and adjust the weights in combination with the output result of the breed inference unit.

[0049] As a further solution of the present invention, the health scoring unit is configured to perform the following operations: Construct a health assessment index system for domestic pets during the growth and development period, including the body fat percentage index, body fluid distribution index, and tissue metabolism index; Based on historical data, construct an association model between the assessment indexes, and incorporate the animal category, breed classification, and the output result of the breed inference unit as influencing factors into the model; Train a health scoring model using the XGBoost integrated learning method; Input the real-time monitoring data into the health scoring model and calculate the growth and development health score.

[0050] The hyperparameters of the XGBoost model are set as follows: learning rate 0.05, tree depth 8, minimum number of samples in a leaf 10, regularization term λ = 1.0, and the input features include the body fat percentage index, body fluid distribution index, tissue metabolism index, and breed inference adjustment factor.

[0051] The database management module in the embodiments of the present invention includes: an individual information table for storing the basic information and breed information of domestic pets at different growth and development stages; an appearance feature table for storing the feature vectors of standard breeds and the appearance feature data of pets of unknown breeds input by users; a clustering model table, a measurement data table, and an analysis result table; a hair compensation coefficient library for storing hair compensation coefficients under different hair lengths and densities.

[0052] The user interface in the embodiments of the present invention includes: an animal information input area, including a known breed selection mode and an unknown breed feature input mode. The known breed mode is used to select the animal category, specific breed, and input appearance features. The unknown breed mode is used to input appearance features such as body size, hair characteristics, ear shape, and tail characteristics; a hair status selection area for selecting the current hair status of the pet, including the shaved state and the hairy state. When the hairy state is selected, the hair length grade and the estimated hair density can be further input; a real-time data display area for displaying the values of body fat percentage indicators, body fluid distribution indicators, tissue metabolism indicators, and the breed similarity analysis results of pets of unknown breeds; a result analysis area for displaying the body composition change trends, standard reference ranges, and growth and development health scores at four growth and development stages.

[0053] Example 1. In a pet clinic environment, a veterinarian uses a portable animal body composition analyzer to analyze the body composition of a 3-year-old known breed golden retriever. The analysis process is as follows: The veterinarian first starts the portable animal body composition analyzer and enters the user interface. In the animal information input area, the veterinarian selects the "known breed selection mode", further selects the animal category as "dog", the specific breed as "golden retriever", and inputs the basic information of the pet, including the name "Lucky", age "3 years old", weight "32 kg", and gender "male".

[0054] Next, in the hair status selection area, the veterinarian selects the current hair status of the pet as the "hairy state", and further inputs the hair length grade as "grade 3" (medium-length hair) and the estimated hair density as "0.8" (relatively dense). This information is recorded in the individual information table and the appearance feature table of the database management module.

[0055] After the breed inference unit receives the input pet appearance feature data, since it is a known breed, the system directly retrieves the standard feature vector of the golden retriever from the database and calculates the breed inference adjustment factor α = 1.0 (fully matching the known breed).

[0056] Subsequently, the veterinarian conducts data collection. First, the veterinarian uses a bioelectrical impedance device and a near-infrared spectroscopy device to measure the core measurement points. According to the anatomical landmark diagram displayed by the device, the veterinarian sequentially places the measurement probes at the midline of the pet's neck, the center of the chest, the center of the abdomen, and the center of the back. To improve the measurement accuracy, the veterinarian also measures the auxiliary measurement points, including the points on both sides of the back and the distal points of the front and hind limbs.

[0057] After the raw data is collected by the data collection module, these data are transmitted to the data processing module. The data processing module filters out high-frequency noise from the bioelectrical impedance data, uses the wavelet transform method to remove random noise from the near-infrared spectroscopy data, and performs dynamic baseline correction through polynomial fitting. Finally, the data is subjected to maximum-minimum normalization processing to make the data from different sources comparable.

[0058] The processed data is transmitted to the dynamic correction unit. The dynamic correction unit retrieves the previous measurement records of this dog from the historical database as a reference and constructs a time series model based on the improved LSTM network. This model comprehensively considers the historical body composition data and the breed inference adjustment factor to generate an error compensation value.

[0059] Since the pet is in a hairy state, the dynamic correction unit also retrieves the hair compensation coefficient library data from the database management module. According to the input hair length grade "level 3" and hair density "0.8", the system calculates the hair state compensation coefficient and the dynamic compensation value. Finally, the original measurement value is corrected to the corrected measurement value through a formula.

[0060] The corrected data is transmitted to the body composition analysis unit. The body composition analysis unit extracts the impedance characteristic parameters of each measurement point from the bioelectrical impedance data to construct a body fat distribution feature vector; extracts the spectral features related to proteins from the near-infrared spectroscopy data to construct a tissue metabolism feature vector; and extracts the features related to water from the near-infrared spectroscopy data to construct a body fluid distribution feature vector. Subsequently, the system uses a weighted average fusion method to perform multi-modal data fusion on these three feature vectors and finally calculates the body fat percentage, body fluid distribution index, and tissue metabolism index.

[0061] Next, the health scoring unit receives the output results of the body composition analysis unit. Based on the health scoring model trained by the XGBoost ensemble learning method, the system calculates the growth and development health score of this golden retriever.

[0062] Finally, the real-time data display area of the user interface shows the values of the body fat percentage, body fluid distribution, and tissue metabolism index. The result analysis area then displays the body composition change trend graph, the standard reference range, and the growth and development health score of this dog in the adult stage. The veterinarian can view these data and provide health advice to the pet owner based on the results.

[0063] After the entire analysis process is completed, the system stores the data measured this time in the measurement data table and the analysis result table of the database management module for future reference as historical data.

[0064] Example 2. This example introduces the pheromone-assisted technology, which specifically addresses the problems of unstable pet activities and low measurement cooperation in the home environment. The pheromone-assisted technology is implemented through an integrated or detachable pheromone diffusion module, which can release low concentrations of pet calming pheromones (such as feline pheromone component F3, canine DAP pheromone, etc.) before measurement to reduce the stress response of animals and improve stability during the measurement process.

[0065] The working mechanism of the pheromone diffusion module is as follows: Before the device is started, the user can choose whether to enable the pheromone diffusion function and its parameter settings; After the device is started, the system controls the pheromone release amount according to the preset process, causing it to diffuse during a predetermined time before measurement and continue to release until the measurement ends; The pheromone diffusion module uses a low-power fan diffusion method to ensure uniform release without interfering with the measurement accuracy of the device.

[0066] This technology is particularly suitable for the home environment or dealing with easily frightened animals, which can effectively reduce the stress response, improve the measurement success rate and data stability. The system supports the user to select different types of pheromone preparations according to the animal species to match the individual needs of different animals.

[0067] In the home environment, the user uses a portable animal body composition analyzer equipped with this technology to perform body composition analysis on a juvenile feline of unknown breed. The specific operation process is as follows: The user takes out the portable animal body composition analyzer and observes that the device has an integrated pheromone diffusion module. Considering the particularity of home environment measurement, the user enables this function, loads the pheromone preparation suitable for felines, and selects "Enable pheromone diffusion function" in the system settings and sets appropriate diffusion parameters.

[0068] Subsequently, the user starts the device and enters the user interface. In the animal information input area, select the "Unknown breed characteristic input mode", specify the animal category as "cat", and input the basic information of the animal, including the identification name, age "8 months", measured weight, and gender information. At this time, the pheromone diffusion module starts to operate, and the system controls the pheromone to be released according to the preset program.

[0069] The user inputs the appearance feature data of felines according to the interface guidance, including body size, body proportion, hair length level, hair color, ear shape, and tail characteristics. In the hair state selection area, the user designates the current hair state as "hairy state" and inputs the corresponding hair parameters.

[0070] The breed inference unit receives the input appearance feature data, converts it into a feature vector, compares it with the known feline breed feature vectors in the database, and calculates the similarity. The system detects that the similarity with a certain breed is lower than the preset threshold, so a clustering algorithm is used to divide the body type categories and calculate the corresponding breed inference adjustment factor.

[0071] After the pheromone diffusion module works for the preset time, the system prompts that the effective concentration has been reached. The user starts data collection and uses a bioelectrical impedance device and a near-infrared spectroscopy device to collect data at the measurement points of the feline. Under the action of the pheromone, the stress response of the animal is reduced and remains relatively stable, enabling the user to complete the data collection at the core measurement points and also collect data at some auxiliary measurement points.

[0072] After the data collection is completed, the collected data is transmitted to the data processing module. The system records the relevant information on the use of pheromone assistance during the measurement process. The data processing module performs noise filtering, dynamic baseline correction, and normalization on the bioelectrical impedance data and near-infrared spectroscopy data. The system evaluates the stability and quality of the data and confirms that it meets the processing standards.

[0073] The processed data is transmitted to the dynamic correction unit. Since it is the first measurement and there is a lack of historical data of this animal, the system retrieves the standard development data of animals of the same type and the same age group from the database as a reference. The system generates an error compensation value applicable to juvenile felines based on a time series model.

[0074] Considering that the animal is in a hairy state, the dynamic correction unit retrieves the hair compensation coefficient data from the database, combines the input hair parameters and the breed inference adjustment factor, calculates the hair state compensation coefficient and the dynamic compensation value, and corrects the original measurement value.

[0075] The corrected data is transmitted to the body composition analysis unit. This unit extracts the characteristic parameters from the various data, constructs the corresponding feature vector, and calculates the body fat percentage, body fluid distribution index, and tissue metabolism index through a multi-modal data fusion method.

[0076] The health scoring unit conducts a health assessment based on the results of the body composition analysis. The system takes into account that the animal is in the growth and development stage and adjusts the weights according to the output results of the breed inference unit, calculates the growth and development health score, and provides a health status assessment and corresponding suggestions.

[0077] The user interface displays the analysis results, including the body composition index values, the breed similarity analysis results, as well as the comparison with the standard reference range and the health score. The system also provides an assessment of the use of the pheromone-assisted function, indicating the role of this function in improving the measurement stability.

[0078] After the measurement is completed, the system stores the data in the database, creates an individual record for the animal, records the measurement data and analysis results, providing basic data for subsequent monitoring.

[0079] The above is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

[0080] Finally: The above is only the preferred embodiment of the present invention and is not used to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention shall all be included within the protection scope of the present invention.

Claims

1. A portable animal body composition analyzer, comprising a data acquisition module, a data processing module, a database management module and a user interface, characterized in that, It further includes an algorithm analysis module. Among them, the data acquisition module is connected to the data processing module, the data processing module is connected to the algorithm analysis module, the algorithm analysis module is bidirectionally connected to the database management module, the algorithm analysis module is connected to the user interface, and the database management module is connected to the user interface; The data acquisition module is used to collect the original data of the body composition of domestic pets from the bioelectrical impedance device and the near-infrared spectroscopy device, and the bioelectrical impedance device and the near-infrared spectroscopy device are configured as portable devices; The algorithm analysis module is used to analyze and calculate the preprocessed bioelectrical impedance data and near-infrared spectroscopy data, including a breed inference unit, a dynamic correction unit, a body composition analysis unit, and a health scoring unit. Among them: the breed inference unit generates breed similarity information and breed inference adjustment factors based on the appearance feature data; The dynamic correction unit is used to perform dynamic correction on the body composition data of domestic pets by using a time series model based on historical data and combining the breed inference adjustment factor, and is also used to calculate the hair state compensation coefficient according to the hair state information; The body composition analysis unit is configured to receive the corrected data and analyze the body fat distribution, tissue metabolism, and body fluid distribution indicators of domestic pets through a multimodal data fusion method; the health scoring unit is configured to evaluate the health status of domestic pets during the growth and development period based on an integrated learning supervision model and calculate the growth and development health score of an individual.

2. The portable animal body composition analyzer according to claim 1, characterized in that The measurement points of the data acquisition module are divided into core measurement points and auxiliary measurement points. The core measurement points include the midline of the neck, the center of the chest, the center of the abdomen, the center of the back, and the characteristic measurement points determined according to the animal species. During measurement, the positioning is based on the anatomical landmark points; the auxiliary measurement points include the points on both sides of the back and the distal points of the limbs, and their positions are determined according to the relative positions of the core measurement points; after the data acquisition module obtains the core measurement point data, the preliminary analysis can be completed, and the auxiliary measurement point data is collected when higher accuracy is required.

3. A portable animal body composition analyzer according to claim 1, characterized in that, The breed inference unit is configured to perform the following operations: Receive the pet appearance feature data input by the user through the user interface, and the appearance features include body size, body proportion, hair length, hair color, ear shape, and tail characteristics; Convert the appearance feature data into a feature vector, compare it with the known breed feature vector library in the database, and calculate the similarity; Judge the processing path according to the similarity. When the similarity is higher than the preset threshold, a random forest classification algorithm is used for breed classification; when the similarity is lower than the preset threshold, a K-Means clustering algorithm is used for body type category division; the training data of the random forest classification algorithm comes from the appearance feature vectors in the known breed database; Calculate the breed inference adjustment factor according to the result of breed classification or body type category division for the calculation of subsequent processing units.

4. A portable animal body composition analyzer according to claim 1, characterized in that, The dynamic correction unit is configured to perform the following operations: Establish a historical database to store the body composition measurement data of domestic pets at different growth and development stages. The measurement data includes bioelectrical impedance data and near-infrared spectroscopy data reflecting the body surface fat content, protein content, and water content; Perform time series analysis on the historical data, extract the characteristic change patterns of each growth and development stage, and construct an error compensation model; Input the data of each measurement point collected into the error compensation model to generate the expected parameter values of body fat content, protein content, and moisture content; Compare the expected parameter values of body fat content, protein content, and moisture content with the actual measured values, and calculate the error values of each index; Construct a compensation function for correction based on the error value and the breed inference adjustment factor output by the breed inference unit. The compensation function considers the relationship between the error vector, breed standard weight, and actual weight; Apply the compensation function to the real-time data of the measurement point to generate the corrected numerical values of each index, and store the correction result in the historical database.

5. The portable animal body composition analyzer according to claim 1, characterized in that, The dynamic correction unit is further configured to perform the following operations: Receive the hair status identifier input by the user through the user interface. The hair status identifier is used to indicate whether the pet is in a shaved state or a hairy state currently; When the hair status identifier is in the hairy state, retrieve the data of the hair compensation coefficient library from the database management module; Calculate the hair condition compensation coefficient and the dynamic compensation value based on the breed inference adjustment factor output by the breed inference unit and the hair length grade and hair density characteristic value in the appearance feature data; correct the body composition data Revise to: ； In the formula, is the original measurement value, is the hair state compensation coefficient, is the dynamic compensation value, is the breed inference adjustment factor; Output the corrected body composition data to the body composition analysis unit for subsequent analysis.

6. The portable animal body composition analyzer according to claim 1, wherein The body composition analysis unit is configured to perform the following operations: Receive the corrected data and the original data output by the dynamic correction unit; Extract the impedance characteristic parameters of each measurement point from the bioelectrical impedance data and construct a body fat distribution feature vector; Extract the spectral features related to protein from the near-infrared spectral data and construct a tissue metabolism feature vector; Extract the features related to moisture from the near-infrared spectral data and construct a body fluid distribution feature vector; Perform multi-modal data fusion on the three obtained feature vectors by using a fusion method of weighted average or feature splicing. The fusion process combines each feature vector through weight coefficients; Calculate the body fat percentage index, body fluid distribution index, and tissue metabolism index based on the fused features and input them into the subsequent unit.

7. A portable animal body composition analyzer according to claim 1, wherein The health scoring unit is configured to perform the following operations: Construct a health assessment index system for the growth and development period of domestic pets, including the body fat percentage index, body fluid distribution index, and tissue metabolism index; Based on the historical data, construct an association model between the assessment indexes, and incorporate the animal category, breed classification, and the output result of the breed inference unit as influencing factors into the model; Train a health scoring model by using the XGBoost ensemble learning method; Input the real-time monitoring data into the health scoring model to calculate the growth and development health score.

8. A portable animal body composition analyzer according to claim 1, characterized in that, The data processing module is configured to perform noise filtering, dynamic baseline correction, and normalization processing on the collected original bioelectrical impedance data and near-infrared spectral data.

9. The portable animal body composition analyzer according to claim 1, wherein The database management module includes: an individual information table for storing the basic information and breed information of domestic pets at different growth and development stages; an appearance feature table for storing the feature vectors of standard breeds and the appearance feature data of unknown breed pets input by the user; a clustering model table, a measurement data table, and an analysis result table; a hair compensation coefficient library for storing the hair compensation coefficients under different hair lengths and densities.

10. The portable animal body composition analyzer according to claim 1, characterized in that, The user interface includes: an animal information input area, which includes a known breed selection mode and an unknown breed characteristic input mode. The known breed mode is used to select the animal category, specific breed, and input appearance characteristics. The unknown breed mode is used to input appearance characteristics such as body size, hair characteristics, ear shape, and tail characteristics; a hair status selection area, which is used to select the current hair status of the pet, including the shaved status and the hairy status. When the hairy status is selected, the hair length grade and estimated hair density can be further input; a real-time data display area, which is used to display the values of body fat percentage indicators, body fluid distribution indicators, tissue metabolism indicators, and the breed similarity analysis results of unknown breed pets; a result analysis area, which is used to display the body composition change trends, standard reference ranges, and growth and development health scores in four growth and development stages.

Images