176results about How to "Improve comparability" patented technology

The invention discloses a fermentation strain yield prediction method based on artificial intelligence, particularly relates to the field of biological fermentation production, and is used for solving the problems that in the prior art, due to the fact that the yield is difficult to directly map in concentration prediction, monitoring indexes are normal, the final yield deviates from expectation, and reliable strain comparison and production management decision cannot be supported. A historical batch process observation sequence and a material account sequence are aligned to generate a conversion efficiency characterization sequence to form a yield learning sample training yield prediction model, and the conversion efficiency characterization sequence is generated on batches and input into the model to obtain a yield prediction value. And calculating an accounting volume trajectory consistency parameter and a conversion contribution steady-state proportion parameter, inputting the parameters into a coefficient analysis model to obtain a decision credible coefficient, generating a yield risk judgment result for a yield prediction value based on the decision credible coefficient, and realizing a new-batch online prediction record.

The invention relates to the technical field of automatic driving, in particular to an open type automatic driving safety management system and method supporting multi-benefit-party strategy arbitration, and the system comprises an open type safety strategy interface module which is used for defining a standardized strategy description specification; the strategy plug-in management module is used for receiving and loading the strategy plug-ins conforming to the strategy description specifications; the perception and fusion module is used for collecting the vehicle state data, the environment perception data and the service context data and fusing the data to generate a unified context model; the multi-strategy evaluation module is used for screening a candidate strategy set according to a unified context model and a trigger condition when an abnormal event is detected, and evaluating and selecting an optimal strategy; and the strategy execution and monitoring module is used for executing the action template corresponding to the optimal strategy. According to the invention, the automatic driving vehicle can select adaptive degradation strategies according to different operation scenes and business requirements, and the adaptability of the system to diversified business scenes is improved.

The invention discloses a blue-green algae identification and quantification method and device based on unmanned aerial vehicle remote sensing images and deep learning and a medium, and relates to the technical field of information data processing. Combining meteorological data and water quality monitoring data to construct an adaptive dynamic environment algorithm to extract EXIF metadata including camera parameters and attitude angle information, and establishing a geometric projection model to perform coarse orthographic correction on an original aerial image; the method comprises the following steps: extracting a multi-scale cyanobacterial bloom image feature map by stages based on a ResNet architecture and in combination with a feature pyramid network FPN fused with an attention mechanism, and obtaining an instance mask of cyanobacterial bloom through an anchor-free region proposal network, ROI Align and a head network; based on an improved GIS space projection and deep learning algorithm, carrying out high-precision area measurement and calculation on a binary mask image converted from the instance mask; according to the method, the influence of irrelevant interference on blue-green algae identification can be reduced, and the accuracy and comparability of blue-green algae identification and area calculation are improved.

This invention discloses a method for measuring the grounding resistance of transmission towers based on multi-circuit high-frequency synchronous excitation, belonging to the field of power equipment testing and grounding resistance measurement technology. The method is implemented using a multi-circuit high-frequency signal excitation transmission device. First, a multi-circuit measurement calibration parameter library is established, a non-contact coupling unit is installed and the measurement parameters are initialized, then multiple synchronous high-frequency excitation signals are generated, amplified and dynamically impedance matched, and then non-contactly coupled to the grounding down conductor. Subsequently, multiple feedback signals are synchronously acquired and signal characteristics are extracted. Based on the feedback signals, multi-circuit synchronous closed-loop calibration is performed, the grounding impedance and grounding resistance measurements of each circuit are calculated, and anomalies are eliminated and weighted fusion is performed on the multiple measurement values ​​to obtain the final grounding resistance value. This invention eliminates the need to split the grounding down conductor, improving measurement efficiency, synchronization stability, and anti-interference capability.

The application relates to the technical field of bimetal part manufacturing, and particularly discloses an aluminum-magnesium interface intermediate layer high-flux screening device and method for quantitatively delivering magnesium droplets, which comprises an inert atmosphere experiment cavity; a bearing platform arranged in the cavity and used for fixing an aluminum alloy plate, wherein the bottom of the bearing platform is connected with a two-dimensional moving mechanism and is provided with a heating element; a wire-powder coaxial composite device, which comprises a laser emitter, a beam deflection mirror, a focusing lens and a plurality of independently controlled powder device slots, and a laser beam acts on the surface of the aluminum alloy plate after passing through the focusing lens; a magnesium wire spool and a wire feeding roller, which feeds the magnesium wire from the magnesium wire spool to the heating area of the laser emitter. The device uses a plurality of powder device slots to sequentially deposit a plurality of intermediate layer coatings on the same aluminum alloy plate, and realizes the quantitative delivery of magnesium droplets by combining the quantitative wire feeding and laser heating, so that a large number of aluminum-magnesium interface preparations of different intermediate layers can be completed on the same substrate in parallel, and the screening efficiency and data comparability are effectively improved.

The invention provides a post-retinal detachment reduction body position management method and system and wearable equipment, and relates to the technical field of medical monitoring. The method comprises the following steps: continuously acquiring head motion data through an inertial measurement unit integrated on head-mounted equipment, resolving the data through an embedded processor to obtain a real-time head posture angle, and comparing the real-time head posture angle with a preset rehabilitation position angle range to generate a posture comparison result; meanwhile, neck strain signals are collected through a flexible strain sensor array, and biomechanical characteristic signals are obtained through signal conditioning; and finally, the two types of data are fused, and a judgment result of the consistency of the overall body position of the user is generated through a classification algorithm, so that multi-modal data fusion analysis is realized, the postoperative body position state can be accurately monitored in real time, immediate feedback is provided, and the accuracy of body position management and patient compliance are effectively improved.

The present application relates to a kind of slurry shield slurry chamber pressure control experimental system and method, slurry shield slurry chamber pressure control experimental system physical execution layer includes air cushion chamber, pressure maintaining system and circulation system, pressure maintaining system is used to adjust air cushion chamber pressure, and circulation system is adjusted by adjusting slurry amount, slurry amount adjustment air cushion chamber liquid level;Upper computer monitoring layer includes virtual slurry chamber module and slurry chamber pressure control module, virtual slurry chamber module simulates the change of virtual slurry chamber pressure, and slurry chamber pressure control module is used to carry the slurry chamber pressure control algorithm to be tested.The present application constructs experimental system by "real and virtual combination", avoids the cost and safety risk existing in field experiment, guarantees that a large number of, sufficient, even destructive algorithm test can be safely carried out under the condition of limited laboratory;It can solve the technical problems that new type control algorithm with slurry chamber pressure as direct control target lacks sufficient, reliable test verification platform.

The invention relates to the technical field of medical informatics, in particular to a pain management-oriented physiological signal real-time monitoring and intervention system, which comprises a determination module for determining the abnormal tendency of a stimulation window updated in real time at each updating moment to obtain an abnormal tendency sequence; the determination module is also used for determining similar fluctuation segments in the abnormal tendency sequence; the determination module is also used for determining an abnormal process factor of the physiological sensing signal at the current moment based on the abnormal tendency difference value and the length of each similar fluctuation section; the determination module is also used for determining a trigger factor for evaluating the abnormal pain state at the current moment according to the abnormal process factors of each physiological sensing signal at the current moment and the previous moment, and performing normalization processing on the trigger factor to obtain a normalized trigger factor; and the intervention module is used for carrying out visualization and intervention on the pain degree of the target user based on the normalized trigger factor. According to the invention, the physiological signal abnormity identification accuracy and sensitivity are improved.

The invention discloses an S-type pitot tube calibration coefficient prediction method based on CFD, and the method is realized through the following steps: S1, carrying out CFD numerical simulation based on parametric modeling, and obtaining a calibration coefficient set of an S-type pitot tube under different installation angle deviations, flow velocity range conditions and environmental parameters; s2, integrating simulation data, and constructing a feature vector data set of the key influence parameters; s3, dividing the data set into a training set and a test set; s3, based on an RBF algorithm, constructing a high-precision nonlinear prediction model by using the training set; and S4, verifying the performance of the model by using the test set, and judging whether the prediction accuracy meets the expectation or not. A CFD simulation technology and a machine learning algorithm are integrated, the problem that a traditional real flow calibration device is limited in capacity and cannot reproduce on-site complex conditions is solved, and a solution is provided for high-precision prediction and correction of calibration coefficients under on-site complex installation conditions and a non-uniform flow field.

The application provides a high-temperature alloy forge piece long-time organization stability analysis method, comprising the following steps: sampling from a high-temperature alloy forge piece to obtain an alloy forge piece sample; the alloy forge piece sample is heat treated to obtain a solid solution aging high-temperature alloy test rod; the solid solution aging high-temperature alloy test rod is subjected to long-term aging to obtain an alloy long-term aging test rod; the alloy long-term aging test rod is processed to obtain an alloy long-time mechanical property test rod and an alloy long-time metallographic sample; the alloy long-time mechanical property test rod is subjected to mechanical property analysis; the alloy long-time metallographic sample is subjected to metallographic organization analysis; the correlation between the delta phase precipitation caused by alloy long-time organization degradation and the mechanical property is analyzed; and whether the alloy forge piece meets the long-term use requirements of the required high-performance high-temperature alloy forge piece is predicted. The analysis method of the application can predict whether the high-temperature alloy forge piece meets the long-term use requirements of the required high-performance high-temperature alloy forge piece of an aero-engine, a gas turbine and the like.

The invention discloses a photoelectric sensor system conversion efficiency calibration system and method based on a reverberation chamber, and belongs to the technical field of photoelectric sensor conversion efficiency calibration, the system comprises an electric wave reverberation chamber, a transmitting antenna, a reference antenna and a photoelectric field sensor system; the photoelectric field sensor system is composed of a photoelectric sensor, a laser source, a photoelectric detector, a polarization maintaining optical fiber and a single-mode optical fiber. In the calibration process, the reference antenna and the photoelectric sensor system are used as receiving ends respectively, receiving power at different stirring positions in the reverberation chamber is sampled to obtain two groups of transfer functions, and the total efficiency of the photoelectric sensor system is determined by comparing the two groups of transfer functions under the condition that the radiation efficiency of the reference antenna is known. The conversion efficiency of the photoelectric sensor system is obtained by combining the relation among the power, the field intensity and the total efficiency in the free space; and the effective averaging of the incident direction and the polarization state is realized by using the uniform and isotropic electromagnetic field characteristics of the reverberation chamber.

The application discloses an analysis method for silicon and manganese alloy element enrichment behavior of hot-rolled plate coil oxide scale, and belongs to the technical field of hot-rolled plate coil surface oxide scale detection and evaluation, comprising: sampling, preparing a sample for a test section, identifying the hierarchical structure of the oxide scale, the oxide scale-matrix interface and the abnormal area, determining the measurement path and / or measurement area across the thickness direction of the oxide scale and the oxide scale-matrix interface; screening and positioning silicon (Si) and manganese (Mn) in the test section, quantitatively performing line scanning and / or area scanning on the test section by using EPMA to obtain concentration distribution data of target elements; establishing a coordinate system, determining the reference concentration of the target elements, and calculating the enrichment index according to the concentration distribution data; comparing the enrichment index with a pre-established pickling defect risk threshold value, and outputting the result. The application solves the problems of insufficient authenticity of the test section, non-uniform measurement path, difficult comparison of the results and insufficient stability of the determination basis in the prior art.

The invention discloses a financial analysis report generation method, and relates to the technical field of natural language processing and information extraction, and the method comprises the steps: enabling natural language analysis to be a core position before generation, carrying out the differential analysis of a criterion text, and extracting the constraints such as a disclosure item, a metering caliber, a period, a unit and currency; then, structured and normalized alignment with multi-source heterogeneous financial data is completed, so that the response of the system to external criterion change does not depend on a large amount of instant annotation or frequent whole-model fine tuning any more; on the basis of a consistency verification score and an adaptive threshold value calculated by a constraint graph, rigid violation and tolerable deviation are distinguished and can be quantitatively positioned, and all-link backtracking of evidence anchor points is matched to ensure that each paragraph and each conclusion have audible sources, so that compliance risks caused by outdated calibers and disclosure missing are remarkably reduced.

The invention relates to the technical field of gas detection, in particular to a gas recognition device, a gas recognition method and an electronic nose system.The gas recognition device comprises a constant-temperature gas circuit unit, a sensor array, a calibration unit and a data processing unit; the constant-temperature gas circuit unit is used for receiving to-be-tested gas and providing a temperature-controllable testing environment for the to-be-tested gas; the sensor array is used for acquiring information of to-be-detected gas and converting the information into an electric signal; the calibration unit is connected with the data processing unit and is used for generating a reference signal, transmitting the reference signal to the data processing unit and providing a calibration reference for the electric signal of the sensor array; the data processing unit is connected with the constant-temperature gas circuit unit and is also connected with the sensor array; and the data processing unit is used for controlling the formation and switching of the test environment, receiving the electric signal and identifying the information of the gas to be tested.

The invention discloses a heating and ventilation outdoor unit health degree prediction method based on infrared imaging of an unmanned aerial vehicle, and particularly relates to the technical field of heating and ventilation outdoor unit health degree prediction. Comprising the following steps: acquiring a first infrared temperature frame and a second infrared temperature frame which are continuously acquired according to an air route in the same inspection task, and a first visible light frame and a second visible light frame which are respectively aligned in time, and outputting an inspection frame set; based on two frames of visible light, an outdoor unit start-stop state, a fan gear state and a spraying partition start-stop state are extracted to form a two-time operation element set, consistency verification is carried out with the condition that the three states are consistent one by one as a contrast establishment condition, a supplementary collection mark is output and recollection is triggered if the condition does not meet the condition, and a contrast failure mark is still not met if the preset number of times of recollection is exceeded. The output contrast frame pair is satisfied; the method comprises the following steps: constructing a contrast frame pair which meets the requirement of consistent operation elements in the same inspection task, and executing combined processing of eliminating operation disturbance terms and writing degradation deviations into digital twin state records on partition response differences.

The application provides a three-dimensional reconstruction and quantitative analysis consistency control method and system of multi-center CT blood vessel imaging data, an electronic device and a storage medium, and relates to the technical field of medical image processing. The method comprises the following steps: first, synchronously acquiring original projection data of multi-center CT blood vessel imaging, blood vessel center line topological data and blood vessel respiratory motion displacement data, generating a blood vessel deformation correction matrix based on the respiratory motion displacement data, and dynamically registering the blood vessel center line data by using the matrix to obtain corrected center line data; then, inputting the corrected center line data and the original projection data into a multi-modal feature fusion network for three-dimensional reconstruction to generate an initial blood vessel model; then, performing morphological parameter quantitative analysis on the initial model; finally, comparing the analysis result with a clinical standard and adjusting the model parameters to output a target blood vessel three-dimensional model. The application improves the geometric accuracy and clinical applicability of the blood vessel model reconstruction.

The invention discloses an under-crown temperature inversion method and system and application, and relates to the technical field of forest ecology and microclimate inversion, and the under-crown temperature inversion method comprises the following steps: obtaining flux tower observation data; constructing and training a baseline prediction model to obtain a baseline prediction value of the vertical temperature difference; constructing and training a gating correction function based on the stability index; adding the baseline predicted value and the output value of the gating correction function to obtain a final predicted value of the vertical temperature difference; and obtaining the under-crown temperature through inversion by subtracting the final predicted value of the vertical temperature difference from the on-crown temperature. The invention provides a flux tower scene-oriented under-crown temperature inversion method and system and application, and the method comprises the steps: reconstructing a vertical temperature difference through employing the conventional environment elements of a flux tower and carrying out the inversion of the under-crown temperature under the condition of lacking of the on-crown / under-crown surface temperature or lacking of temperature profile observation; therefore, the availability, comparability and interpretability of the microclimate data of the site are improved.

The application discloses a kind of flow state solidified soil complex space filling performance evaluation method and intelligent detection device, belong to geotechnical engineering material performance detection and underground engineering construction technical field, evaluation method is through setting standardization test condition, make the flow state solidified soil to be measured in constant liquid level driving and low disturbance release in turn through standardization straight line restricted flow area, path turning area, obstacle flow area and blind area simulation area, collect key parameters in flowing process and calculate multidimensional evaluation index, and then obtain standardization comprehensive filling index CFIs and carry out grade division;Engineering adaptation comprehensive index CFIp can also be calculated in combination with target engineering scene parameter and output construction optimization suggestion.Intelligent detection device includes standard condition setting unit, feeding release unit, constant liquid level feeding unit, standardization flow channel unit, sensing acquisition unit, data processing unit and result output unit.The application realizes the unity of the quantification of evaluation result, the comparability and the engineering guidance.

This invention discloses a method and system for embryo quality detection, relating to the field of embryo detection technology, including the following steps: S1: fertilized embryos are continuously cultured in an in vitro culture system containing specific metabolic marker detection components, and continuous dynamic imaging of the embryos is performed to obtain a complete dynamic developmental image sequence; S2: the dynamic developmental image sequence is segmented, and temporal developmental features are captured, identifying key time points such as pronucleus appearance and disappearance, cleavage, and blastocyst formation, and generating a segmentation mask corresponding to each frame of the dynamic developmental image sequence; this invention achieves full-process time-difference imaging to capture the temporal features of embryonic development, without trauma or interference, avoiding subjective bias in static observation; by segmenting the dynamic developmental image sequence and capturing temporal developmental features, the segmentation accuracy is high and the anti-interference ability is strong.

The application relates to the technical field of active sonar detection, in particular to a transmission waveform selection method for target wake field disturbance detection, which comprises the following steps: applying uniform engineering constraints to selected engineering realizable waveforms to construct a candidate transmission waveform set of a detection system; equivalent the influence of target wake field disturbance on sound propagation to a time delay perturbation process of multi-path propagation, and establishing a target wake field disturbance sensitivity model accordingly; using the target wake field disturbance sensitivity model, calculating the sensitivity indexes of each candidate transmission waveform in the candidate transmission waveform set and sorting, taking the candidate transmission waveform with the largest sensitivity index as the optimal transmission waveform most suitable for target wake field detection; instructing the transmission end of the detection system to transmit a sound wave signal based on the optimal transmission waveform. The method can improve the observability of weak propagation time-varying characteristics caused by wake disturbance, and make the selection of the transmission waveform have stronger explainability and scene adaptation ability.

The application belongs to the technical field of computer hardware testing, and discloses a multi-sensor integrated detection dock system for notebook computer performance testing, which comprises a multi-sensor module, a data acquisition and interface module, a main control module and a hardware component. The application has the following remarkable advantages: high integration and improved efficiency; data synchronization precision and improved analysis value; high automation degree and reduction of human error; consistent test conditions and strong result comparability.

The present application relates to a tire rubber performance detection method, and particularly relates to a method and system for measuring a low-temperature network dissipation index for characterizing the low-temperature rolling energy consumption tendency of rubber tire tread rubber. The present application performs a strain amplitude scanning under constant frequency at T0+ / -DeltaT three temperature points by RPA / DMA, calculates the area type Payne value A(T) of the logarithmic strain coordinate and the temperature derivative thereof at T0, and outputs an index; the mixed area can be normalized or G'' is introduced. Through regression calibration k of the whole tire rolling resistance according to ISO 28580 and a threshold value, fast, comparable and quantifiable prediction of low-temperature energy consumption of the material grade is realized, and the present application is suitable for formula development, incoming material release and process control.

The application discloses a coal-fired unit carbon emission accounting method and system for BECCS, and a medium, and relates to the technical field of carbon emission accounting, which comprises the following steps: obtaining full-chain accounting basic data sets, including fuel basic parameters, emission factor parameters, coal-fired unit operation parameters, supply chain parameters and CCS system parameters; taking a predetermined unified power supply quantity benchmark and an accounting boundary as core benchmarks to determine the benchmark emission of the pure coal combustion benchmark line full life cycle link, the first carbon emission corresponding to the biomass replacement emission reduction link of the BECCS scenario combustion emission, the second carbon emission corresponding to the biomass supply full life cycle link, the third carbon emission corresponding to the raw coal supply chain link in the BECCS scenario, the fourth carbon emission corresponding to the power generation efficiency loss additional link, the fifth carbon emission corresponding to the CO2 capture system full link, the sixth carbon emission corresponding to the CO2 transportation and storage link, and calculating the net carbon emission of the BECCS scenario full life cycle.

The invention discloses a calligraphy stroke feature extraction method and system based on a fixed-size grid. The method relates to the technical field of stroke feature extraction, and comprises the following steps of calligraphy stroke image preprocessing and grid division, grid-based stroke space distribution feature extraction, size feature extraction, grid-based stroke direction feature extraction and feature fusion and output. The method comprises the following steps: S1, normalizing a binary image of calligraphy strokes to a fixed canvas and dividing the binary image into grids; s2, calculating the stroke density of each grid and constructing a density feature vector; S3, refining the image to obtain a skeleton, calculating a direction field and generating a direction feature vector; and finally, the two vectors are connected in series and fused, and a joint feature vector is output for a subsequent task, so that the calligraphy stroke feature extraction accuracy is improved, and the problem of low calligraphy stroke feature extraction accuracy caused by insufficient understanding of dynamic essence, complex form and high-level semantics of calligraphy art in the prior art is solved.

This invention belongs to the field of medical image processing technology, specifically a system for aortic motion feature extraction and disease risk prediction based on medical image analysis. The system includes modules for acquiring multi-temporal image sequences and clinical data, dynamic segmentation of aortic anatomical structures, construction of a dynamic relative coordinate system, trajectory preprocessing and standardization, trajectory segmentation analysis based on physiological phases, multidimensional dynamic feature extraction, and risk stratification prediction model construction. This invention integrates multidimensional motion features with clinical indicators to construct a risk model based on machine learning or statistical regression, outputting a predicted value for the individual's cardiovascular disease risk. This invention can reduce global interference caused by respiratory and cardiac translation, improve the biological specificity and inter-individual comparability of features, achieve fully automated processing, and enhance the risk prediction ability for cardiovascular events such as heart failure, atrial fibrillation, and aortic disease, showing promising application prospects.

The invention discloses a cultivated land non-agrochemical intelligent monitoring method and system based on multi-temporal SAR and optical image fusion, and belongs to the technical field of remote sensing monitoring and land resource management. The method solves the problems of information loss and misjudgment caused by cloud and mist shielding or noise interference of a single remote sensing data source. According to the scheme, the method comprises the following steps: acquiring and preprocessing a multi-temporal SAR and an optical image sequence; carrying out space registration; constructing a double-branch deep neural network, respectively extracting radar scattering features and spectral texture features, and fusing the radar scattering features and the spectral texture features by adopting a cross attention mechanism; generating a change intensity graph based on a time sequence change detection module; and extracting non-agrochemical suspected pattern spots through threshold segmentation, and aggregating and outputting a final monitoring result. The method is mainly used for realizing the automatic and high-precision monitoring of the non-agrochemical change of the cultivated land.

A kind of submarine landslide disaster chain test system and monitoring method for realizing bottom erosion belong to the field of marine engineering, geotechnical disaster physical simulation and test technology, including water tank main module, source item generation module, bottom erosion-shallow buried pipeline response module and multi-source monitoring and synchronous acquisition module.Firstly, the conditions of setting water depth, flow rate and tank body inclination are formed by water tank main body and circulating water supply module;Secondly, the underwater slope sliding or landslide material continuous release boundary is formed by source item generation module, and the landslide material moves along the water tank and then enters the erosion reaction zone, which produces scouring, erosion and bed surface incision effect on the bottom model;At the same time, the overlying water body is disturbed by the movement of landslide material and the process of bottom erosion, and surge response is formed;Finally, synchronous monitoring and unified analysis are carried out by multi-source monitoring and synchronous acquisition module.The present application can realize the test simulation of submarine landslide-bottom erosion damage-shallow buried pipeline response-surge response disaster chain process.

This invention discloses an adaptive thermally excited infrared thermal wave nondestructive testing method and device. The method includes: querying a pre-stored database based on the scanned position signal to obtain the theoretical standard energy distribution spectrum under the current pose; triggering pre-aiming sensing; acquiring and generating the actual pre-aiming energy distribution spectrum; comparing the two to calculate the actual energy non-uniformity, and comparing it with a preset threshold to decide whether to initiate compensation; driving the actuator to perform physical adjustments and confirming the adjustments; and finally synchronously triggering the main thermal excitation and infrared thermal image acquisition. The device includes a scanning motion and pose feedback module, a high-precision multispectral pre-aiming sensing module, an intelligent decision-making and synchronous control module, a finely adjustable main excitation execution module, and a data management and communication interface module. This invention can proactively correct the uneven excitation energy distribution caused by scanning geometric changes from the physical source, significantly improving the quality and reliability of the detected image.

This invention relates to the fields of medical data analysis and artificial intelligence, proposing a dual-latent variable decoupled data analysis method based on plasma Netrin-1 and clinical scale features. First, plasma Netrin-1 test data and clinical characteristics such as UPDRS scale scores and disease duration are acquired from the subjects. Batch effect correction is applied to the raw test values, and an input feature vector is constructed. Then, a variational autoencoder model is established. By setting orthogonal constraints, Netrin-1 regression constraints, and disease duration regression constraints, the latent space is structurally constrained, enabling the model to learn latent variables D representing disease progression and C representing adaptive changes during training. Based on this, an efficiency index η composed of latent variables C and D is calculated, and relevant trend indicator parameters are obtained by combining longitudinal follow-up data. This method can achieve decoupled representation of disease-related features and adaptive change features in multi-source clinical data, providing a new data analysis tool for statistical analysis and research of neurodegenerative disease-related data.