1156 results about "Multi-task learning" patented technology

Provided in the present application is an intelligent question answering method based on a collaboration between a large language model and a knowledge graph, relating to the technical fields of artificial intelligence and natural language processing, the method comprising: decomposing a complex question into a plurality of simple questions, and analyzing the degree of association between the simple questions and a basic function so as to form a multi-hop reasoning path; automatically extracting structured information from the simple questions on the basis of a multi-task learning framework of a large model, so as to construct a knowledge graph; and constructing a cumulative reasoning learning framework on the basis of a logic reasoning large model, and performing iterative verification on a process result formed by the knowledge graph on the basis of the multi-hop reasoning path, so as to correct the reasoning path until a correct answer is inferred.

The invention belongs to the technical field of medical image processing, and discloses a CT image segmentation and classification system based on segmentation feature guidance, and the specific technical scheme is as follows: the system adopts a shared encoder to extract general features, realizes collaborative optimization of segmentation and classification through a double decoding path, adopts a partial decoder in a segmentation path, and adopts a partial decoder in the segmentation path; in combination with a local feature attention module, through multi-scale feature fusion and a boundary perception mechanism, the region consistency of a global segmentation map is gradually optimized, edge detail information is supplemented, and a classification path generates a space attention weight through a segmentation feature guide module by utilizing segmentation prediction; the classification network is guided to focus on a focus area and suppress background interference, a self-adaptive loss weighting strategy based on multi-task learning is adopted, the double-task gradient flow is dynamically balanced, and the gradient competition problem in the multi-task learning is effectively relieved.

The invention discloses an intestinal cancer immune response graph neural network prediction method, a medium and equipment, and the method comprises the steps: collecting pathological image information, immunodetection information and basic clinical information, extracting a tissue space distribution characteristic spectrum through a deep convolutional network, and constructing a graph neural network model in combination with an immunomarker expression characteristic matrix; spatial interaction characteristics of a tumor microenvironment are modeled by adopting a graph attention mechanism, finally a treatment response probability, an optimal treatment opportunity and an adverse reaction risk are predicted through a multi-task learning framework, and a clinical decision report containing a prediction response curve, a risk early warning threshold and a treatment time window suggestion is output. According to the method, through multi-modal data fusion and spatial interaction modeling, accurate prediction of intestinal cancer immunotherapy response is realized, and a more comprehensive reference basis is provided for clinical decision making.

The invention relates to the technical field of power grid monitoring, in particular to a power operation risk identification method, system and device based on multi-modal data fusion and a storage medium. In order to solve the problems of multi-modal data splitting, topological constraint missing and the like in traditional power disturbance analysis, an improved BERT model is constructed, and electrical signal time-frequency features and text semantic information are mapped to a unified vector space through a multi-modal embedding mechanism; a time sequence attention mechanism is adopted to establish a time dependency relationship between signals and texts, and a graph attention network is combined to realize risk propagation modeling under power grid topology constraints; and collaborative optimization of disturbance classification, risk prediction and trend analysis is carried out through a multi-task learning framework. The technical problems that heterogeneous data fusion is difficult and risk identification precision is insufficient are effectively solved, accurate identification and intelligent early warning of electric power operation risks are achieved, and the safe operation level of a power grid is improved.

The invention discloses a power equipment fault prediction system based on big data analysis. The method comprises the following steps: acquiring initial equipment multi-dimensional data; constructing a dynamic topology network of the power equipment, including a dependency relationship between the equipment and a fault propagation path, performing embedded learning on the dynamic topology network by using a GNN graph neural network, and extracting equipment collaboration features in the initial equipment multi-dimensional data; a multi-task learning framework is constructed in combination with the equipment cooperation features to predict the equipment fault probability and the remaining service life, and an equipment health index is obtained; and acquiring environmental parameters, dynamically adjusting a fault judgment threshold based on the equipment health index and the environmental parameters, generating a prediction result, integrating the prediction result with an SCADA system, and triggering graded early warning. And the influence of environmental factors on the operation state of the equipment is fully considered. Under different environmental conditions, the equipment fault risk can be judged timely and accurately.

The invention provides a city calculation basic model and equipment based on a Mama time sequence, and relates to the technical field of data processing. In the model provided by the embodiment of the invention, a Mama framework is combined with multi-task time sequence analysis. Aiming at the non-stationarity of urban time series data, a frequency prompt network is designed for decoupling in combination with a Koopman operator theory and dynamic mode decomposition, a steady mode and non-stationary fluctuation are analyzed from the angles of time and frequency, and a frequency mode memory pool (FPMP) is introduced for realizing cross-scene feature reuse. The advantage of a Mama model in capturing long-term time dependence is utilized, and an attention mechanism similar to Mama is designed to simulate channel dependence in a multivariable time sequence. A learnable supplementary sequence and a multi-task prompt mark are introduced to support dynamic multi-task adaptation, and the contradiction between urban data isomerism and task diversity is relieved. Shared knowledge and specific task knowledge are balanced through shared experts and specific task experts of the multi-task attention module, the negative migration problem in multi-task learning is relieved, and the generalization ability of the model is enhanced.

The invention provides a software supply chain risk analysis method. A data acquisition and preprocessing module acquires component information from a plurality of software warehouses and performs format unified processing; the components are modeled as nodes through a hypergraph construction module, the dependency relationship is modeled as hyperedges, and a hypergraph structure is dynamically constructed and updated so as to express many-to-many dependency relationship more accurately; performing representation learning on the hypergraph structure through a hypergraph neural network, generating embedded vectors of the components, and fusing context features of the components; based on an embedded vector, risk scoring, anomaly detection and classification are carried out by adopting multi-task learning and a graph attention mechanism, so that the generalization ability of the model and the risk identification precision are improved; online incremental updating of a hypergraph structure and model parameters is achieved through a dynamic updating and feedback mechanism, and the real-time performance and stability of the model are ensured. According to the method, the data transmission security, the transmission efficiency and the compatibility between devices are improved, and the method has higher risk analysis capability and system adaptability.

The invention relates to the crossing field of precision agriculture and artificial intelligence technology, in particular to a soil component detection system based on machine learning and infrared spectroscopy, which is characterized in that a soil spectrum is acquired on site through a portable Fourier infrared spectrometer, and after pretreatment, feature vectors are constructed by fusing climate, soil and crop data; the deep learning decision-making module extracts spectral features by using one-dimensional convolution, fuses multi-dimensional information through an attention mechanism, synchronously outputs a fertilization scheme, crop suitability scores and soil improvement measures by a multi-task learning sub-network, and finally, verifies by combining an agronomic knowledge base so as to obtain a fertilization result. According to the method, a comprehensive decision report containing a quantitative fertilization formula, a crop suitability sequence and a soil improvement scheme is generated, precise agricultural guidance is realized, spectral features are automatically extracted by adopting a one-dimensional convolutional neural network, and a multi-modal data fusion and multi-task learning framework is combined, so that the system achieves relatively high precision in the aspect of soil nutrient prediction.

The invention discloses a goaf collapse risk assessment data fusion system based on big data processing, and particularly relates to the technical field of geological disaster assessment, and the system comprises three core modules: a multi-source data adaptive weighted fusion module which establishes a unified space-time coordinate system, converts non-raster data into a continuous field through Kriging interpolation, and performs data fusion on the continuous field; combining the information entropy and the correlation coefficient to dynamically distribute weights, and generating an enhanced feature field through self-supervised pre-training; the physical-space-time neural network dynamic prediction module is integrated with elastic-plastic mechanical constraint loss and multi-task learning, and outputs a future multi-time step risk probability field and a deformation prediction field through a space-time convolution-memory network; and the risk field three-dimensional subdivision and emergency response module is used for clustering three-dimensional voxels in a high-risk area, automatically calculating risk body parameters, generating an emergency scheme in combination with DEM data and an A * algorithm, and improving evaluation accuracy and emergency scheme practical operability through digital twinborn deduction evaluation.

The invention relates to the technical field of detection, in particular to an automatic water quality detection method and device, and the method comprises the steps: carrying out the multi-scale collection processing of a sample water body based on a distributed sensor network, extracting the spatial correlation features of multiple levels in an original data set, carrying out the feature reconstruction of the original data set according to the spatial correlation features, generating a feature enhancement data set; performing graph structuring processing according to the feature enhancement data set, and outputting to obtain a water quality feature graph; inputting the water quality feature map into a multi-task learning model, optimizing the multi-task learning model based on a loss function, and outputting a multi-parameter prediction result; performing reverse propagation processing on the multi-parameter prediction result to generate a spatial distribution diagram of the water quality parameters; and performing multi-scale pollution influence assessment based on the spatial distribution map, and generating a comprehensive water quality monitoring report. The core problem that the overall pollution condition and the dynamic change trend of the water body are difficult to accurately reflect in the prior art is solved, and the precision and practicability of water quality monitoring are remarkably improved.

The invention discloses a farmland field segmentation method and system based on multispectral SAM model and multitask learning guidance, and belongs to the field of field segmentation. The problem that in the prior art, channel number matching is conducted in a simple channel compression or deep convolution mode, and the challenges that the boundaries of cultivated land parcels are complex, diversified and subtle are difficult to effectively deal with is solved. The method comprises the following steps: performing spectrum compression on a multispectral remote sensing image through a U-Net structure fused with a CBAM attention mechanism to generate a three-channel pseudo visible light image; the image is input into an SAM-ViT encoder subjected to LoRA fine tuning, and global semantic features are extracted; a task token sequence is constructed, explicit interaction between the token and the image features is realized through a bidirectional Transform module, and high-resolution fusion features are generated; calculating cross-task attention, and outputting a multi-task mask through inner product operation; and performing post-processing by using a watershed algorithm to generate a field boundary. The method is used in the field of cultivated land identification.

The invention provides a PCBA (Printed Circuit Board Assembly) multi-defect collaborative diagnosis method fusing hypergraph diffusion and cross-modal comparison, which comprises the following steps of: constructing a multi-layer hypergraph structure, and generating a multi-modal defect sample by adopting a condition hypergraph diffusion model based on the multi-layer hypergraph structure. Performing cross-modal feature alignment on the multi-modal defect sample by adopting a multi-layer alignment mechanism, and calculating a defect interaction matrix of aligned features; according to the defect interaction matrix, node features of each node in the node layer are updated by adopting a message passing mechanism of a graph neural network; and inputting the node features into a multi-task learning framework, and predicting the defect of each node to obtain a node defect prediction result. According to the method, complex correlation between PCBA defects can be effectively modeled, multi-mode sensor information is fully utilized, and the method has real-time collaborative diagnosis capability so as to meet the urgent requirements of the modern electronic manufacturing industry for high-precision, high-efficiency and high-reliability quality detection.

The invention relates to the field of geological exploration and geotechnical engineering, provides a rock mass dynamic mechanical parameter sensing method and system based on while-drilling parameters in order to dynamically obtain rock mass dynamic mechanical parameter characteristics in situ in real time in the drilling process, and provides a rock mass dynamic mechanical parameter sensing method and system based on while-drilling parameters by using multi-modal signals such as vibration, acoustic emission, torque and bit pressure through uniform characteristic engineering and multi-task learning. The instantaneous elastic modulus, the dynamic Poisson's ratio, the transient uniaxial compressive strength, the cohesive force and the internal friction angle can be synchronously and dynamically predicted in single drilling, and the limitation that only a single static parameter is estimated in a traditional method is broken through; a rock mass dynamic mechanical parameter acquisition model constructed on the basis of a multi-modal adaptive graph attention sequential network can effectively couple the physical proximity and the statistical association relationship of a sensor, and integrates expansion sequential convolution and a time perception Transform module, so that the characterization capability of emergencies (such as jamming / friction resistance) and long-term trends is considered on the premise of low time delay; therefore, the prediction accuracy is improved.

The invention provides an arch dam shape efficient intelligent optimization method and system based on an agent model. The method comprises the steps that an arch dam physical-numerical model is constructed, a double optimization target with structural safety and economical efficiency as the core is determined, design parameters are selected, a constraint function is set, and an evaluation index system is established; samples are generated through Latin hypercube sampling, a data set is constructed in combination with finite element calculation, and a multi-task learning architecture is adopted to train a high-precision agent model. And then, coupling the proxy model with a multi-objective optimization algorithm, quickly searching a Pareto optimal solution set, and screening out a comprehensive optimal figure by using a multi-attribute decision-making method. And finally, through a finite element simulation verification result, prediction precision and performance improvement are ensured. The method has the advantages of lightweight modeling, efficient prediction and accurate search, and provides an effective tool for intelligent optimization and rapid decision making of hydraulic structures such as arch dams and the like.

The invention discloses a system and a method for predicting postoperative recurrence risk after triple negative breast cancer neoadjuvant therapy based on multi-modal time sequence medical image data, and belongs to the field of medical image analysis. The system comprises a data processing module used for constructing a multi-modal data set; the multi-modal feature extraction and screening module is used for extracting deep learning, radiomics and tumor habitat features from the region and carrying out feature screening; the model training module is used for constructing a time sequence model based on a Transform architecture and carrying out training through a multi-task learning strategy integrated with time consistency constraint and gene association auxiliary loss; and the recurrence risk prediction module is used for loading the trained model and outputting a recurrence probability and a risk level. According to the method, the multi-modal time sequence image and gene information are fused, so that the recurrence risk of the triple negative breast cancer patient is dynamically and accurately quantified, and support is provided for clinical individualized treatment decision.

The invention relates to the technical field of sewage detection, provides an intelligent detection method for suspended particles in coking sewage based on image deep learning, and aims to solve the problems of detection lag and low precision in the prior art. According to the technical scheme, the method comprises the following steps: synchronously acquiring sound wave and image data through a high-frequency ultrasonic transducer and an industrial camera, and extracting acoustic parameters and optical characteristic parameters; using a dual-channel convolutional neural network to fuse features, and combining a cross-modal attention mechanism to perform dynamic weighting; and outputting a particle concentration classification result and a particle size distribution prediction value through a multi-task learning model, and performing online learning to compensate environmental interference factors. The device is mainly used for realizing real-time and high-precision suspended particle monitoring and automatic process adjustment of coking sewage treatment.

The invention relates to the field of artificial intelligence and big data processing, and discloses a professional field agent construction method and system based on a multi-modal hybrid expert model and a knowledge graph. The method comprises the following steps: collecting multi-source heterogeneous data such as user behaviors, texts, images, structured data and voices, and vectorizing and coding the multi-source heterogeneous data; performing feature extraction on different modal data and fusing to form unified representation; constructing a hybrid expert network based on a gating mechanism to perform task specialization processing on the fusion representation; introducing a domain knowledge graph and performing semantic enhancement through a graph neural network; on the basis of fusion representation, tasks such as recommendation, intention recognition and question and answer are executed at the same time through a multi-task learning framework; and an output result is fed back through the multi-modal interaction interface, and user feedback is collected to update the model. According to the method, the semantic understanding, task prediction and personalized recommendation capabilities of the model in a complex scene can be improved.

The invention relates to the technical field of power dispatching management and control, and discloses a power distribution network power dispatching method based on a virtual power plant AI large model and demand response, and the method comprises the steps: collecting the operation data of a power distribution network, and constructing a feature vector; an AI large model is adopted to calculate and predict load output, and joint uncertainty information is output; calculating a system power unbalance amount, and generating a scheduling strategy; issuing a scheduling instruction corresponding to the scheduling strategy and executing the scheduling instruction; comprehensive performance evaluation indexes are calculated, and whether a performance reduction reason diagnosis mechanism is started or not is judged; according to the method, the AI large model is adopted, the load power, the photovoltaic output and the wind power output are predicted at the same time through a multi-task learning strategy, and the correlation among multiple variables is fully utilized; by constructing a multi-objective optimization model, comprehensively considering economy, safety and reliability and adopting an improved particle swarm optimization algorithm for solving, coordinated optimization configuration of demand response resources is realized, power grid fluctuation is effectively reduced, and power supply reliability is improved.

The invention discloses an AI-driven disease diagnosis and curative effect monitoring analysis system, which comprises a data processing module used for collecting multi-source diagnosis and treatment data and constructing a multi-modal sample set; the diagnosis prediction module is used for generating a diagnosis label, a diagnosis confidence value and a prediction curative effect trend through a multi-task learning model; the curative effect comparison module is used for collecting actual curative effect data, constructing a curative effect observation sequence and generating a curative effect deviation sequence based on the curative effect observation sequence and the predicted curative effect trend; the credibility evaluation module is used for executing credibility review based on the curative effect deviation sequence and the diagnosis confidence value; and the path evolution module is used for recording continuous multi-round diagnosis correction results, generating a jump type diagnosis path and updating a diagnosis and treatment data chain. The dynamic closed-loop verification mechanism between the multi-mode diagnosis and treatment data and the predicted curative effect trend improves the accuracy of disease diagnosis, the reliability of curative effect prediction and the adaptability of the diagnosis and treatment process.

The invention discloses a crop disease and pest identification method and system based on multi-task learning. The method comprises the following steps: firstly, constructing a crop disease and pest identification model comprising a multi-scale feature information fusion module, a disease detection branch and a severity classification branch; the method comprises the following steps: preprocessing an input leaf image, and extracting and fusing a multi-scale feature map by a multi-scale feature information fusion module; the disease detection branch generates disease and insect pest candidate regions by using a region proposal network, and outputs disease and insect pest positions and categories through disease detection in combination with a multi-scale fusion feature map; meanwhile, the fusion feature map with the maximum size is input into a severity classification branch to realize four-stage evaluation; a weighted loss function design thought is provided, and multi-task network branches can be guided to carry out joint training. According to the method, end-to-end multi-task cooperative processing is realized, disease and pest positioning, classification and severity evaluation are synchronously completed by sharing a multi-scale fusion feature map, the recognition efficiency is greatly improved, and the real-time monitoring requirement of an agricultural scene is met.

The invention discloses a sewage treatment plant effluent prediction method based on multi-task learning. The method comprises the following steps: acquiring sewage treatment data; based on the sewage treatment data, establishing an effluent prediction model; the input of the effluent prediction model is inflow water quality data, process data, environmental data and sewage treatment unit data, and the output of the effluent prediction model is predicted effluent index data; predicting the water outlet index data in future time based on the water outlet prediction model and the input of the water outlet prediction model; according to the method, the water outlet prediction model is constructed in combination with multi-task learning, and the model can comprehensively consider the time sequence dependence and mutual influence relationship among the input data to perform prediction, so that the calculation redundancy is reduced, and the prediction demand that an actual process needs to cooperatively consider multiple targets is met; the effluent quality prediction precision and the process regulation and control efficiency are remarkably improved, and a solid foundation is laid for promoting intelligence of operation management of a sewage treatment plant.

The invention relates to a medical training analog simulation system based on big data analysis, in particular to the field of medical training analog simulation, the medical training analog simulation system provides an efficient, accurate and dynamically adaptive training environment through cooperative work of a plurality of modules; the data processing module ensures high quality of input data, the construction module improves the generalization ability of the model through transfer learning and multi-task learning, the dynamic optimization module improves the model prediction precision through self-supervised learning and learning rate adjustment, and the real-time feedback module monitors student operation and provides personalized tutoring; the whole system combines high-quality data, intelligent training and real-time feedback, provides accurate and personalized support for medical education and practical training, and improves operation skills and coping ability of students.

The present invention relates to the technical field of natural ventilation in buildings, and in particular to a computationally assisted decision-making method and system for climate-adaptive building cavity design. The method comprises: using a Delaunay triangulation method to generate an initial mesh, using a Laplace operator-based mesh refinement method to adaptively refine the initial mesh, so as to obtain an adaptive mesh system, wherein the adaptive mesh system is used for dynamically adjusting the mesh density of conditional PINNs; constructing a multi-task learning framework within the conditional PINNs, wherein the multi-task learning framework is used for jointly predicting a plurality of physical field variables within the conditional PINNs; and iteratively training the conditional PINNs, and using a gradient descent algorithm to minimize a loss function until a predetermined number of training iterations or loss convergence is reached, thereby generating a key physical field variable prediction model in building cavity design. The present invention can implement efficient and accurate prediction of key physical quantities in building cavity design, and can be adapted to different building layouts and functional space characteristics.

Techniques are disclosed for training a model, using multi-task learning, to transform natural language to a logical form. In one particular aspect, a method includes accessing a first set of utterances that have non-follow-up utterances and a second set of utterances that have initial utterances and associated one or more follow-up utterances and training a model for translating an utterance to a logical form. The training is a joint training process that includes calculating a first loss for a first semantic parsing task based on one or more non-follow-up utterances from the first set of utterances, calculating a second loss for a second semantic parsing task based on one or more initial utterances and associated one or more follow-up utterances from the second set of utterances, combining the first and second losses to obtain a final loss, and updating model parameters of the model based on the final loss.

The invention provides a software vulnerability understanding method based on adversarial training data enhancement and multi-task learning, belongs to the technical field of software quality assurance, and solves the technical problems that vulnerability type prediction and row-level vulnerability positioning are difficult to accurately complete at the same time and code semantics are easy to damage due to sample enhancement in the existing method. According to the technical scheme, the method comprises the following steps: (1) constructing a C / C + + vulnerability data set containing vulnerability type tags and row-level vulnerability tags; (2) generating an adversarial training data enhancement method constrained by AST and PDG in the embedding space by adopting a PGD algorithm; and (3) taking the pre-training model as a shared encoder. The method has the advantages that the performance of the two tasks can be remarkably improved, and the manual workload needed in the vulnerability understanding and analysis process is effectively reduced.

The invention provides an intelligent prediction model construction method based on multi-modal data fusion, and the method comprises the following steps: collecting multi-modal data in different electric power scenes, and carrying out the preprocessing of the multi-modal data, the electric power scenes including electric power inspection, electric power transaction and electric power equipment operation and maintenance; performing feature extraction, semantic association and space-time alignment on the preprocessed multi-modal data through a cross-modal attention mechanism and association analysis; constructing an intelligent prediction model, and performing knowledge injection on the intelligent prediction model through a retrieval enhancement generation technology and a lightweight adapter; a multi-task learning framework is introduced into the intelligent prediction model, exclusive output heads are set for different power scenes, and training is carried out through multi-modal data; after training is completed, lightweight compression is carried out on the intelligent prediction model through a knowledge distillation technology, a multi-modal data fusion technology system suitable for the power industry is formed, and key technology support is provided for deepening application of the intelligent prediction model of the power system.

The invention relates to the technical field of remote sensing data processing and earth surface parameter inversion, and discloses an earth surface matrix parameter hyperspectral data inversion method and system based on machine learning. Comprising the following steps: constructing a multi-source heterogeneous hyperspectral data set; performing feature screening on the preprocessed hyperspectral data set based on an adaptive band selection algorithm, constructing a dynamic weight matrix by calculating mutual information entropy and inter-class distance measurement between spectral bands to realize intelligent screening of key feature bands, and combining spectral derivative conversion and spectral index calculation to generate an enhanced feature vector; and a multi-task transfer learning neural network model is constructed, and an output layer realizes multi-parameter collaborative inversion based on a multi-task learning architecture. And performing preprocessing and feature enhancement operation which is the same as that of the training data on the hyperspectral image data of the to-be-inverted region, inputting the trained neural network model, and outputting a surface matrix parameter inversion result.

A crop genotype-to-phenotype general prediction method based on multi-task learning comprises the following steps: collecting genotype and phenotype data sets of five crops, performing quality control and coding on genotype data, and performing standardization processing on phenotype data; dividing the data set into a training set and a test set in proportion by adopting a stratified sampling method; by taking genotype data as input and phenotype data as output, a universal prediction model based on a multi-task learning framework is constructed, information sharing among different phenotype prediction tasks is realized, and potential relations between complex phenotypes and genotypes are accurately captured. According to the method, phenotypes can be accurately predicted according to genotypes in the early stage of breeding, so that materials with target characters are rapidly screened, the breeding period is effectively shortened, the breeding efficiency is remarkably improved, and an efficient and universal technical scheme is provided for crop breeding and related research.

The invention discloses a blockage prediction and self-adaptive regulation and control method for a forging press cleaning system. The method comprises the following steps: carrying out multi-sensor data acquisition based on a multi-sensor fusion mode; preprocessing the multi-sensor data to obtain a standardized data set; a deep learning model fusing the attention mechanism and multi-task learning is constructed, and the deep learning model takes an LSTM network as a basic framework and comprises a feature extraction module, an attention enhancement module and a multi-task learning module; training and optimizing the constructed deep learning model based on training data input; predicting a blockage trend, inputting the real-time fusion feature vector into a trained model, and outputting probability distribution of a blockage state and a blockage risk level by the model; and performing self-adaptive regulation and control, namely generating a self-adaptive regulation and control instruction, and controlling an execution mechanism to execute fan power adjustment or reverse blowing mode switching operation. The invention further discloses a corresponding system, electronic equipment and a computer readable storage medium.

The invention discloses a spine image key point detection algorithm based on multi-task learning. The algorithm comprises the following steps: constructing a multi-task deep learning network model comprising a segmentation branch and a key point positioning branch; a feature aggregation module based on multi-scale cavity convolution is embedded in the segmented branches, and the multi-scale cone feature extraction capability of the model is enhanced through splicing fusion of multiple cavity rate convolution branches and global pooling branches; a cross-task attention fusion module is introduced between the two branches, and bidirectional dynamic interaction and complementation between segmentation features and key point features are realized by generating and fusing first-order and second-order context attention maps; semantic alignment loss is designed in a training stage, collaborative optimization of two tasks is promoted by constraining the consistency of segmentation masks and key point heat maps in a high-level feature space, global information of segmentation and local information of key point detection are fully utilized, and the accuracy and stability of spine centrum key point positioning and the accuracy of a segmentation result are improved.