Patents
Literature
Patsnap Eureka AI that helps you search prior art, draft patents, and assess FTO risks, powered by patent and scientific literature data.

611results about How to "Improve personalization" patented technology

Cerebral hemorrhage personalized treatment scheme optimization method and system based on big data analysis

InactiveCN119153117ARealize multi-criteria dynamic optimizationensure safetyMedical data miningDrug and medicationsPatient characteristicsTreatment options
The invention provides a cerebral hemorrhage personalized treatment scheme optimization method and system based on big data analysis, and relates to the technical field of data analysis, and the method comprises the steps: constructing a cerebral hemorrhage patient multi-source heterogeneous clinical database, extracting feature subsets, carrying out semantic alignment, carrying out time sequence modeling and fusion, and obtaining a patient panoramic feature representation; constructing a cerebral hemorrhage diagnosis and treatment knowledge graph, adding the cerebral hemorrhage diagnosis and treatment knowledge graph to a preset graph attention network, distributing weights, obtaining knowledge-enhanced patient feature representation and taking the knowledge-enhanced patient feature representation as a state, generating an action space based on an optional treatment scheme, constructing a reward function and determining a candidate treatment scheme; performing medication suitability analysis, performing semantic conflict identification in combination with the patient allergy history and complication, and performing screening to obtain a suitable candidate treatment scheme set; and for each scheme, predicting a multi-dimensional prognosis index, constructing a multi-criterion decision matrix, combining personal influence factors, adjusting the weight of each index, modeling an optimization problem and determining an optimal treatment scheme.
Owner:DONGGUAN SOUTHEAST CENTRAL HOSPITAL (DONGGUAN SOUTHEAST TRADITIONAL CHINESE MEDICINE MEDICAL SERVICE CENTER DONGGUAN FIRST HOSPITAL AFFILIATED TO GUANGDONG MEDICAL UNIVERSITY)

Association rule mining-based collocation recommendation method and system

The invention provides a collocation recommendation method and system based on association rule mining, and relates to the technical field of association rule mining, and the method comprises the steps: obtaining user historical transaction data, generating a user commodity transaction score matrix, carrying out the standardization processing, and predicting the commodity score through calculating the cosine similarity between users, and obtaining an initial candidate set; carrying out normalization processing on commodity prices, calculating commodity comprehensive similarity in combination with a category hierarchical structure, carrying out clustering grouping, and screening frequent item sets meeting support degree and confidence threshold values; and finally, a commodity price change sequence is extracted to calculate a dynamic price matching degree, a multi-dimensional evaluation index is constructed, a non-dominated sorting genetic algorithm and a deep Q network are adopted for optimization, and scene-based dynamic evaluation and recommendation are realized. According to the invention, the accuracy and timeliness of commodity collocation recommendation are improved, and the interpretability of the recommendation result is enhanced.
Owner:QINSILK COM

Standard substance and standard substance retrieval and sorting method and system based on search engine

The invention provides a search engine-based standard substance and standard substance retrieval and sorting method, which relates to the technical field of data processing, and comprises the following steps: acquiring retrieval keywords input by a user, constructing a standard substance and standard substance knowledge base, mapping to a low-dimensional semantic space, calculating the semantic correlation between the retrieval keywords and the standard substance knowledge, and constructing a heterogeneous information network. Generating comprehensive relevancy measurement, and performing multi-granularity semantic matching to obtain candidate retrieval results; for structural features corresponding to the candidate retrieval results, adaptive feature weights are distributed, independent clustering is carried out, subspace clustering results are obtained, optimization is carried out, global optimal clustering results are obtained, high-order semantic association features are mined, cross-category semantic association is carried out, and high-quality retrieval results are generated; performing low-rank decomposition to obtain an initial comprehensive score, modeling a sorting problem as a Markov decision process, determining an action space and a state space, constructing a reward function, determining a sorting position, and obtaining an optimal sorting result.
Owner:TAN-MO TECH CO LTD

Digital exhibition hall multimedia equipment intelligent interaction control method and system

The invention relates to the technical field of multimedia interaction, in particular to an intelligent interaction control method and system for digital exhibition hall multimedia equipment, and the method comprises the following steps: S1, carrying out the arrangement of a sensor, and carrying out the intelligent sensing and data collection; s2, generating an intelligent interaction strategy; s3, equipment cooperative control and execution; and S4, performing real-time feedback and system optimization. According to the method, user behaviors and environment data are collected in real time through the multi-modal sensor, user requirements are accurately speculated in combination with a deep learning algorithm, the optimal interaction strategy and task priority are generated, individuation and interaction efficiency of user experience are remarkably improved, multi-device cooperation is optimized through distributed task scheduling, and user experience is improved. According to the method, display content and environment parameters are dynamically adjusted, self-optimization of the system is realized by utilizing user feedback, real-time performance, smoothness and reliability of interactive experience of the exhibition hall are ensured, rapid fault response and accurate maintenance demand prediction are realized through anomaly detection and equipment log analysis, and the stability and sustainability of operation of the exhibition hall system are greatly improved.
Owner:ZHEJIANG KUAIBU CULTURE TECH CO LTD

Efficient flight control system of low-altitude manned unmanned aerial vehicle based on tilting composite wings

The invention discloses an efficient flight control system of a low-altitude manned unmanned aerial vehicle based on tilting composite wings, and relates to the technical field of unmanned aerial vehicle control systems. The low-altitude manned unmanned aerial vehicle efficient flight control system based on the tilting composite wings comprises an initialization and pre-check module, a vertical take-off and landing control module, a transition flight control module, an obstacle avoidance and path planning module and a landing control module. Through integration of vertical take-off and landing, smooth transition to cruise, intelligent obstacle avoidance, path planning and other modules, autonomous and efficient flight in a complex environment is realized, and the application scene of the unmanned aerial vehicle is greatly widened. Meanwhile, the accurate flight control and data post-processing analysis capability not only guarantees the safety and comfort of passengers, but also promotes the continuous optimization of the system performance through the real-time performance evaluation and feedback mechanism. In addition, the aerial refueling / charging technology is supported, the cruising ability is prolonged, and the task execution continuity and flexibility are enhanced.
Owner:BEIJING ZHAOYANG SCI TECH CULTURE

Text-driven automatic image generation method and system

The invention discloses a text-driven automatic image generation method and system, and belongs to the technical field of image generation, and the generation method comprises the following specific steps: I, receiving and analyzing text data input by a user, carrying out explicit modeling on entities and relationships in a text, and extracting implicit languages in the text data; according to the method, the multi-modal alignment of the text and the visual features is improved, the visual deviation problem caused by semantic information errors is avoided, the detail richness and semantic information expansion capability of the generated image are improved, the global layout reasonability of the generated image is improved, and the individuation and stylization capabilities of the generated image are enhanced; according to the method, a rigid effect caused by a fixed template is avoided, the performance of the generated image on scene complexity and detail richness is improved, the accuracy of relation modeling between objects in a complex scene is remarkably improved, the intelligent level of image generation is improved, and the generation process is more adaptive.
Owner:SHANGHAI SHINIAN SPORTS TECH CO LTD

Driving skill training method based on diversified virtual examination room construction

The invention discloses a driving skill training method constructed based on a diversified virtual examination room, and relates to the technical field of data processing, and the method comprises the steps: constructing a system based on the diversified virtual examination room, constructing a virtual examination room scene to carry out the driving training of a training user, outputting a driving behavior data set in real time, and carrying out the scoring, and analyzing the data according to the driving score, generating a personalized training path in combination with historical data when the driving score does not reach the standard, and training the user to carry out training step by step according to the training path until a preset skill scoring target is achieved. The technical problems that in the prior art, a traditional driving training method cannot dynamically adjust a training path to meet the actual skill requirements of trainees, consequently, training content lacks pertinence and is low in efficiency are solved, the individuation and pertinence of driving skill training are improved, it is ensured that the training content is highly matched with the actual skill requirements of the trainees, and the driving skill training efficiency is improved. And the training efficiency and effect are improved.
Owner:WUHAN FUTURE MIRAGE TECH CO LTD

Recommendation method and device based on knowledge graph, equipment and storage medium

The invention relates to a recommendation method based on a knowledge graph, and the method comprises the steps: carrying out the semantic analysis of knowledge data of a target domain through a bidirectional LSTM auto-encoder, and generating a semantic vector; in combination with a semantic vector and named entity recognition technology and a semantic role labeling technology, entities and relationships thereof are recognized, a triple is constructed, and a structured knowledge graph is generated. Node feature extraction and embedded representation are performed in a knowledge graph, an initial graph neural network model is constructed in combination with graph structure information, a target graph neural network model is generated through training and optimization, nodes are scored and sorted, and a target object is recommended. According to the invention, the accuracy and intelligent level of recommendation can be improved, the personalization and matching degree of recommendation results are enhanced, and the problem of inaccurate recommendation results in the prior art is significantly improved.
Owner:PING AN TECH (SHENZHEN) CO LTD

Large-model-driven chronic disease follow-up visit and management platform

The invention discloses a chronic disease follow-up visit and management platform driven by a large model, and relates to the technical field of medical information technology platforms. The system is applied to managing and tracking the follow-up visit process of chronic disease patients and providing health consultation and answers, and comprises a user interface module, a logic control module, a database module, a large model module and an auxiliary function module, and the user interface module is used for providing an interactive interface between a patient and a doctor. An intelligent and efficient system platform is constructed by integrating a large model technology, a voice recognition technology and a digital human technology, the system platform adopts a privatized large model, data security and privacy protection are ensured, medical institutions are allowed to autonomously optimize and control the application of the model, and on the whole, through an advanced technical means, the intelligent and efficient system platform has the advantages of being high in practicability and easy to popularize. Refining, individuation and high efficiency of chronic disease management are remarkably improved, and the purpose is to improve the health outcome and life quality of patients.
Owner:SOUTHEAST UNIV

Intelligent pleasure boat navigation system based on interactive pointing control and method thereof

The invention provides an intelligent pleasure boat navigation system based on interactive pointing control and a method thereof. The system comprises a monitoring display module; a single-point navigation instruction receiving module; a route planning module; a multi-point navigation indication processing module; the real-time coordination and broadcast module is a user interaction interface module; the invention further discloses an intelligent pleasure boat navigation method. By adopting an interactive pointing control technology, a dynamic neighborhood potential field path planning algorithm and a multi-target optimization algorithm, the problems of the traditional navigation system in the aspects of safety, efficiency, user experience and multi-ship management are solved, the safety, efficiency and user experience of navigation are remarkably enhanced, the capability of multi-ship collaborative operation is realized, and the system has a wide application prospect. The method is an important progress of intelligent water traffic management in the future.
Owner:青岛无疆技术有限公司

Advertisement recommendation method and system based on click rate prediction model

The invention discloses an advertisement recommendation method and system based on a click rate estimation model, and relates to the technical field of advertisement recommendation, and the method comprises the steps: collecting advertisement recommendation feature data, carrying out the processing through a feature project, and outputting a user feature vector; inputting the user interest vector into a click rate estimation model, training the click rate estimation model through a causal optimized cross entropy loss function, and outputting a click probability value of each advertisement; the click probability value serves as input of a reinforcement learning algorithm, and an advertisement sorting result is generated by dynamically optimizing a multi-target reward function in combination with user interaction behaviors; according to real-time feedback data in the user interaction behavior, updating a user interest partial value of the causal inference model in real time, and correcting user interest vector offset; by updating the user interest partial value in real time, the user behavior change can be quickly responded, and the timeliness of advertisement recommendation and the user experience are improved.
Owner:BEIJING HONGTU XINDA TECH CO LTD +2

Remote monitoring system for medical patients

The invention relates to the technical field of remote medical monitoring, in particular to a medical patient remote monitoring system which comprises a health monitoring module, a disease course prediction module, a physiological regulation module, a medicine management module, a behavior analysis module, a therapy customization module, a dose adjustment module and a system comprehensive management module. According to the invention, accurate physiological monitoring is realized through a data fusion algorithm and a noise filtering technology, the accuracy and reliability of data are ensured, the recurrent neural network model can provide more accurate disease development trend analysis, early diagnosis and timely intervention are facilitated, and the accuracy and reliability of the system are improved. A self-adaptive control algorithm and a biological feedback technology enable a patient to more effectively adjust the physiological state by themselves, the treatment initiative is improved, and the safety and effectiveness of drug treatment are improved through a mode recognition algorithm and drug interaction analysis of drug management. Through deep data mining and personalized treatment strategies of the behavior analysis module and the therapy customization module, a more comprehensive and customized treatment scheme is provided for the patient.
Owner:SHENZHEN XIGUWEI BIOMEDICAL TECH

Advertisement budget allocation method and system based on machine learning and user behaviors

The invention discloses an advertisement budget distribution method and system based on machine learning and user behaviors, and the method comprises the following steps: collecting the historical behavior data of a user, and extracting the feature data of the historical behavior data through feature engineering; constructing a user probability prediction model, and training the model by using the feature data; in an advertisement putting process, collecting real-time behavior data of a user, inputting the real-time behavior data into the trained model, and predicting an advertisement clicking probability of the user and an advertisement commodity purchasing probability of the user; a fusion model is constructed, the fusion model calculates the conversion rate of the user according to the advertisement clicking probability of the user and the advertisement commodity purchasing probability of the user, and calculates the channel conversion rate according to the putting channel; distributing a preliminary advertisement budget according to the historical data of the putting channel; and dynamically adjusting the preliminary advertisement budget of the putting channel based on the channel conversion rate. According to the invention, precise distribution of the advertisement budget can be realized, and the overall conversion effect of the advertisement is improved.
Owner:XIAMEN ZHONGLIAN CENTURY TECH CO LTD

Intelligent rehabilitation treatment management system and method based on big data

The invention discloses an intelligent rehabilitation treatment management system and method based on big data, and relates to the technical field of intelligent management. During operation of the system, rehabilitation-related data of an orthopedic patient is collected, real-time data is obtained through intelligent wearable equipment and a sensor, the pain and fatigue state of the patient is monitored, preprocessing is carried out, and the patient is subjected to rehabilitation treatment; the method comprises the following steps: analyzing collected data, monitoring the rehabilitation progress of a patient in real time, calculating to obtain a comprehensive rehabilitation progress evaluation index Etotal, comparing the comprehensive rehabilitation progress evaluation index Etotal with a preset state threshold value M and a second preset state threshold value N to generate a personalized rehabilitation scheme, and remotely monitoring rehabilitation data of the patient by a doctor so as to obtain a personalized rehabilitation scheme. According to the method, the rehabilitation plan and the training progress of a patient can be checked through a mobile phone APP, the rehabilitation plan, the training progress and real-time feedback are fed back, intelligent reminding is obtained, and through rehabilitation video or voice guidance, the patient can share experience, exchange rehabilitation experience and obtain psychological support and encouragement.
Owner:LEER HEALTH TECH (SHENZHEN) CO LTD

Hospital pharmacy dispensing information intelligent processing method and system

The invention relates to the technical field of medical information, in particular to a hospital pharmacy dispensing information intelligent processing method and system, and the method comprises the following steps: S1, obtaining the basic data of a medicine in real time; s2, acquiring medical record information, current medication conditions and physiological parameters of the patient from a hospital information system; s3, analyzing an existing drug combination, and detecting potential drug incompatibility; s4, a medication scheme of the patient is automatically generated; s5, pushing the medication scheme and the taboo information generated in the step S4 to work terminals of doctors and pharmacists, and providing medication suggestions at the same time; and S6, continuously monitoring the reaction of the patient, and when abnormity is monitored, performing drug adjustment through an adjustment algorithm. According to the system and the method, the medicine data is updated in real time, the patient information is integrated, and incompatibility identification and medication adjustment are carried out by utilizing an intelligent algorithm, so that the safety and the efficiency of hospital pharmaceutical dispensing are remarkably improved, and meanwhile, the individuation and the scientificity of a treatment scheme are optimized.
Owner:GENERAL HOSPITAL OF PLA

Lamp energy efficiency optimization method based on artificial intelligence

The invention relates to the technical field of energy efficiency optimization, and discloses a lamp energy efficiency optimization method based on artificial intelligence, and the method comprises the steps: data collection: carrying out the all-around and multi-dimensional data collection of an environment where a lamp is located through a plurality of sensors. According to the lamp energy efficiency optimization method based on artificial intelligence, in order to make energy utilization efficient, the multi-dimensional environment and lamp operation data are obtained through the data acquisition step, optimization is carried out in combination with the adaptive learning algorithm, lamp energy efficiency control parameters can be adjusted in real time according to different environments, places, time periods and other situations, and the energy efficiency optimization efficiency is improved. The lamp can be kept in the optimal energy efficiency state in various complex and changeable environments, the energy utilization efficiency of the lamp is effectively improved, energy waste is reduced, in the scene-based energy efficiency optimization link, for different application scenes, differentiation strategies are implemented according to respective typical lighting requirements and energy-saving potentials, parameters such as lamp brightness are accurately adjusted, and the energy efficiency of the lamp is improved. And the energy utilization efficiency of the lamp in various scenes is improved.
Owner:FOSHAN CITY LITIAN OPTOELECTRONICS

Esophageal cancer multi-mode treatment effect comprehensive evaluation system

The invention relates to the technical field of medical information, in particular to an esophageal cancer multi-mode treatment effect comprehensive evaluation system which comprises a patient data acquisition module, a drug resistance dynamic monitoring and adjusting module, an immune system response dynamic analysis module and a treatment effect evaluation module. The patient data acquisition module is used for acquiring multi-modal health data of a patient; the drug resistance dynamic monitoring and adjusting module analyzes drug resistance signs appearing in the treatment process; the immune system response dynamic analysis module comprehensively evaluates the state change of the immune system; and the treatment effect evaluation module evaluates the treatment effect of the patient in real time based on the drug resistance and immune system response analysis result, and provides treatment adjustment suggestions for the doctor according to the evaluation result. According to the invention, the accuracy of treatment is improved, the risk of treatment failure is reduced, and decision support based on data is provided for doctors.
Owner:JIANGSU PROVINCE HOSPITAL (THE FIRST AFFILIATED HOSPITAL OF NANJING MEDICAL UNIVERSITY) +1

User information input method and system in intelligent television all-in-one machine

The invention relates to the technical field of television image communication, in particular to a user information input method and system in an intelligent television all-in-one machine. The method comprises the following steps: collecting a user image, carrying out white balance correction and focusing processing, generating a moving object focusing image, carrying out identification and attention tendency analysis on a character in the focusing image, obtaining character tendency data, analyzing a historical interaction image, identifying an interaction mode and constructing a user intention prediction model, and obtaining a user intention prediction result. And when the tendency data shows that the user preferences the television content, intention drilling and auxiliary adjustment are performed on the current image based on the prediction model, predicted user input information is generated, logic conflicts in the predicted information are detected, feedback optimization is performed, and final user input information is output. According to the invention, a more accurate and more flexible user information input method is realized.
Owner:SHENZHEN MEIJIA TECH CO LTD

System and method for generating official document based on large model

The invention discloses an official document generation system and method based on a large model, belongs to the technical field of artificial intelligence, and aims to solve the technical problems of low official document format and content accuracy, poor material failure and poor content innovation in official document generation. Comprising a model training module used for performing model training on an official document generation model based on official document samples in an official document sample library; the model fine tuning module is used for performing layered fine tuning on the trained official document generation model; the cue word optimization module is used for carrying out cue word matching on user input based on a built-in cue word set; the outline generation module is used for calling a final official document generation model to generate an official document outline based on the cue word combination and user input; and the official document generation module is used for calling the final official document generation model to generate and output an official document body based on the user input and the final official document outline, and calling the privacy data shielding module to shield the privacy data in the official document body.
Owner:SHANDONG LANGCHAO YUNTOU INFORMATION TECH CO LTD

Patient rehabilitation monitoring method and system based on nursing data

The invention discloses a patient rehabilitation monitoring method and system based on nursing data, and particularly relates to the technical field of rehabilitation monitoring. Core rehabilitation indexes are extracted by acquiring and classifying nursing data of a patient; analyzing the nursing data, identifying the current rehabilitation stage of the patient and evaluating the matching degree of the rehabilitation target and the core rehabilitation index; performing trend analysis on the time sequence of the nursing data, identifying abnormal data and evaluating the interference degree on the rehabilitation process; judging whether the rehabilitation progress of the patient meets the expectation or not according to the evaluation results of the matching degree and the interference degree; when the rehabilitation progress does not conform to the expectation, analyzing the demand change of the patient and evaluating the help effect of the nursing scheme; by comprehensively analyzing the matching degree, the interference degree and the nursing effect, the rehabilitation effect of the patient is evaluated, whether the nursing scheme needs to be adjusted or not is judged, individuation, accuracy and effectiveness of the nursing scheme are improved, the rehabilitation process of the patient can be accelerated, and the nursing quality is improved.
Owner:THE THIRD AFFILIATED CLINICAL HOSPITAL OF CHANGCHUN UNIV OF TRADITIONAL CHINESE MEDICINE

Oral cavity three-dimensional model construction and tooth occlusion characteristic analysis method

The invention relates to the technical field of computer image processing, in particular to an oral cavity three-dimensional model construction and tooth occlusion feature analysis method. The method comprises the following steps: carrying out imaging scanning processing and three-dimensional model construction on an oral cavity structure of a patient to generate an oral cavity internal structure three-dimensional space model; performing tooth segmentation and tooth feature analysis on the oral cavity internal structure three-dimensional space model, and performing inter-tooth contact evaluation analysis to obtain oral cavity inter-tooth arrangement contact condition data; performing tooth occlusal point positioning analysis and tooth occlusal surface connection division on the oral cavity internal structure three-dimensional space model to generate an oral cavity tooth occlusal connection surface; and performing tooth occlusion characteristic analysis on the oral tooth occlusion connection surface to obtain a tooth occlusion surface parallelism and a tooth occlusion surface contour shape, and performing tooth occlusion adjustment processing to generate an oral tooth occlusion surface adjustment scheme. According to the invention, rapid reconstruction of an oral cavity structure and accurate analysis of tooth occlusion characteristics can be realized.
Owner:成都中教智汇信息技术有限公司

Diabetic diet recommendation method and system based on deep reinforcement learning and oriented to multi-objective optimization

The invention discloses a multi-objective optimization-oriented diabetic diet recommendation method and system based on deep reinforcement learning. The method comprises the steps of obtaining a diet data set; extracting user health target features and user diet mode features in the diet data set by using a BERT model; extracting user diet preference characteristics in the user diet interaction data by using a VIT model; constructing a fitness function about the user health target, the user diet preference and the user diet mode, and obtaining an initial diet recommendation scheme by using a Q-MOPSO algorithm in combination with the extracted user health target features, the extracted user diet preference features and the extracted user diet mode features; and taking the initial diet recommendation scheme as a state function of deep reinforcement learning, and performing diabetic diet personalized recommendation through a deep reinforcement learning model, the deep reinforcement learning model comprising a user health prediction module. According to the method, the accuracy, individuation, real-time performance and dynamic performance of diabetic diet recommendation are improved.
Owner:INNER MONGOLIA UNIV OF TECH

Intelligent hospital guide and tracking return visit recording integrated system for medical examination

The invention discloses an intelligent hospital guide and tracking return visit recording integrated system for medical examination, which relates to the technical field of intelligent medical treatment, and comprises the following steps: based on a comprehensive data set, using an A-star algorithm to identify a hospital guide path; in the process that the patient waits for the hospital according to the hospital guide path, the system collects behavior data of the patient in real time; based on historical data, training a waiting time difference prediction model, inputting the patient behavior data into the waiting time difference prediction model, and predicting a waiting time deviation value; dynamically adjusting a hospital guide path according to the waiting time deviation value; and after the patient finishes visiting according to the adjusted hospital guide path, the system automatically provides health propaganda and education contents, collects patient feedback data and stores the patient feedback data in the tracking return visit system. Through the A-star algorithm and the waiting time prediction model, the system can perceive waiting delay in advance and adjust a path in time, the waiting time of a patient is remarkably shortened, and the doctor-seeing experience is improved.
Owner:THE 923RD HOSPITAL OF THE CHINESE PEOPLES LIBERATION ARMY JOINT LOGISTICS SUPPORT FORCE

Method and system for realizing personalized recommendation based on multi-feature fusion face recognition

The invention relates to a method and a system for realizing personalized recommendation based on multi-feature fusion face recognition, and aims to provide real-time, efficient and accurate recommendation service, the method comprises the following steps: collecting a face image of a target audience in real time through a camera device, and preprocessing the image to extract a face region; running a face recognition algorithm in the edge computing device, and extracting multi-modal features including age, gender, appearance and style features; a comprehensive feature vector is generated by using a Transform model through a feature fusion module, and style classification and appearance classification are realized; according to the method, user feature vectors and a content database are matched on the basis of cosine similarity, personalized recommendation content is generated, the weight of a classification model is dynamically optimized in combination with user feedback, and dynamic behaviors and appearance features of a user are accurately captured through a time sequence analysis module, a dynamic style analysis module and a deep appearance analysis module; and comprehensively generating user feature description by combining vision, voice and character multi-mode information.
Owner:BEIJING HUAYUAN TECHNOLOGY CO LTD

Transdermal electrical stimulation equipment based on large language model

The invention discloses a transcutaneous electrical stimulation device based on a large language model, and relates to the technical field of wearable devices. The transcutaneous electrical stimulation equipment based on the large language model comprises a data acquisition and preprocessing module, a feature extraction and fusion module, a model training and prediction module and a real-time information processing and response module, real-time data processing and intervention capabilities are integrated, human body data can be analyzed in real time, electrical stimulation parameters can be automatically adjusted, an instant treatment or adjustment scheme is provided for the problems of arrhythmia, chronic pain and the like, and the timeliness and effectiveness of treatment are greatly improved. And secondly, by fusing the physiological signal features and the symptom text features, the equipment can more comprehensively understand the health condition of the user, so that a more accurate treatment scheme is provided. Besides, the equipment also has the functions of anomaly detection and trend analysis, can predict potential health risks, and triggers an alarm in time when a serious anomaly is detected, thereby effectively guaranteeing the life safety of the user.
Owner:BEIJING WENZHI MEDICAL TECHNOLOGY CO LTD

Discriminator knowledge distillation personalized federated learning method based on conditional generative adversarial network

The invention relates to the technical field of federated learning, and discloses a discriminator knowledge distillation personalized federated learning method based on a conditional generative adversarial network. The problems of appearance of massive participants and influence of non-independent identical distribution of data on federal learning data cost and personalized learning efficiency are solved; a discriminator knowledge distillation model fused with the conditional generative adversarial network is put forward to be embedded into a personalized federated learning framework; the differentiated feature representation of the client model is enhanced through the discriminator knowledge distillation model, so that the personalized model effectively recognizes and captures specific modes and features of individuals, the representation capability and the personalized effect of the model are enhanced, the higher model compression rate is achieved, and meanwhile, the user experience is improved. The conditional generative adversarial network fused with the discriminator generates synthetic data specific to each client, and the synthetic data improves the generalization ability of personalized federal learning and the adaptability of the model on the premise of fully ensuring the data privacy.
Owner:HARBIN UNIV OF SCI & TECH

Personalized federal learning-based drug purchase quantity prediction method

The invention belongs to the technical field of medicine supply, and particularly relates to a personalized federal learning-based medicine purchase quantity prediction method, which realizes model data sharing on the basis of protecting the problem of user privacy disclosure. The drug purchase quantity prediction method comprises the following steps: S1, determining hospitals and drugstores participating in a region, and collecting and preprocessing data; s2, selecting and constructing a federal learning model, wherein the constructed model is a DCNN-LSTM hybrid model; s3, performing training, parameter cutting and transmission, aggregation and optimization on the model constructed in the step S2, wherein the data is not out of the local; s4, performing prediction and application on the optimization model obtained in the step S3; and S5, supply chain adjustment. According to the method, doctor-patient information leakage is prevented, information communication overhead is reduced, development of a medical intelligent system is promoted, and key support is provided for promoting sustainable development of the intelligent medical system and further development of the artificial intelligence industry by utilizing the federated learning technology and accurately predicting the drug demand quantity.
Owner:JIANGXI CHANGHE PHARM SUPPLY CHAIN MANAGEMENT CO LTD

AI-driven multi-dimensional short video matrix analysis and management system

The invention provides an AI-driven multi-dimensional short video matrix analysis and management system, and relates to the technical field of information prediction.The AI-driven multi-dimensional short video matrix analysis and management system is characterized in that audience modal information when a short video is played is obtained by performing audience verification on a traffic feature set, and click attribute information when a short video user clicks the short video is determined according to the audience modal information; determining audience acceptability of the short video content according to the access like data, and determining a delivery matching coefficient during short video delivery according to the audience acceptability and the click attribute information; according to the behavior characteristic information, determining a perception monster degree when the short video user watches the playing of the short video; and performing trend analysis on the dynamic interest tag of the short video user according to the delivery matching coefficient and the user behavior sequence subjected to the correction constraint, so as to perform short video recommendation on the short video user. According to the method, the user experience and the platform activity degree can be improved under the influence that dynamic interests and complex preferences of the user are difficult to accurately capture from the recommended content.
Owner:GUANGZHOU GOODIDEA INFORMATION TECH CO LTD

Intelligent recognition method of vehicle driver fatigue state based on multi-modal physiological information fusion

The present application discloses a method for intelligently identifying the fatigue status of an in-vehicle driver by integrating multi-modal physiological information. The method comprises the following steps: continuously recording a facial video of the driver to obtain M continuous frames of normal images and M continuous frames of fatigue images, using an adaptive illumination equalization technique to adjust the brightness and contrast of the fatigue image to obtain a second fatigue image, using a facial key point detection algorithm to process the normal image and the second fatigue image respectively to obtain normal feature points and fatigue feature points, extracting the normal physiological features and fatigue physiological features of the driver based on the normal feature points and the fatigue feature points, using the normal physiological features and the fatigue physiological features as training data to train a fatigue detection model for outputting a fatigue index, using the fatigue detection model to obtain a fatigue index, and when the fatigue index exceeds a dangerous threshold, issuing a prompt warning to the driver, thereby improving the accuracy and personalization of intelligent identification of the driver's fatigue status.
Owner:HANGZHOU DOT TECH CO LTD

Medical text data synthesis method based on role driving

The invention discloses a medical text data synthesis method based on role driving, and relates to the technical field of text data synthesis and artificial intelligence, and the method comprises the following steps: S100, generating an original problem; s200, screening original problems; and S300, synthesizing the medical text data. According to role description and interest content generation, the adaptability to different medical role requirements is enhanced, semantic duplicate removal and correlation evaluation are realized by utilizing a text coding model and a vector database, semantic duplicate content is removed, correlation is accurately evaluated by utilizing a large language model, the data quality and diversity are improved, and the method is suitable for popularization and application. Through generation of variant questions and answers, diversity and practicability of medical text data are enriched. According to the method, the personalization and pertinence of the medical text data are improved, the quality and duplicate removal efficiency of the medical text data are enhanced, the problem types are enriched, the generated content is optimized, and various information requirements of the medical text data are comprehensively covered.
Owner:NINGBO BODEN AI TECHNOLOGY CO LTD