51 results about "Keywords diabetes" patented technology

The invention discloses a heterogeneous double-flow fusion method and system for diabetic retinopathy grading. The method comprises the following steps: obtaining an output result of diabetic retinopathy grading by utilizing a heterogeneous double-flow architecture; processing an input fundus image into images with different resolutions; extracting global context features from the low-resolution image by using a lightweight visual Transform model distilled by composite knowledge, and extracting local focus features from the high-resolution image by using a convolutional neural network model; performing interactive fusion on the global context features and the local focus features of the double-branch architecture through a symmetric bidirectional cross attention fusion module to obtain enhanced fusion feature representation; and finally, inputting the fusion features into a classifier, and outputting a severity grading result of the lesion. The method aims at improving the accuracy and robustness of hierarchical diagnosis through deep analysis of global information and local details, and can be applied to the medical fields of clinical computer-aided diagnosis, eye image analysis and the like.

The invention provides a gestational diabetes risk prediction system and method based on multi-modal data fusion, and the system comprises a data collection module which is used for integrating clinical indexes and medical record text data; the data preprocessing module converts the multimode data into numerical values and text variables which can be used for modeling; the variable screening module is used for extracting data features by adopting LASSO regression in combination with a recursive feature elimination algorithm and a Clinical-BERT model; the data prediction module is used for constructing a GDM risk prediction model through a dual-channel calculation unit and a fusion unit, generating an accurate risk probability and providing an interpretable clinical index in combination with an SHAP value; and a prediction result is output through the output module in the forms of a dynamic column diagram, a webpage calculator and an API interface, so that clinical operation and application are facilitated. According to the method, the limitation of a traditional prediction method is broken through, the prediction precision and the real-time monitoring capability are improved, and the development of precise medical treatment is promoted.

The invention discloses a 3D holographic image knowledge base interactive playing method for diabetic education, and the method comprises the steps: firstly obtaining patient data of a hospital HIS system and a wearable device, and carrying out the classification, preprocessing and data fusion, thereby obtaining an encrypted standardized data set; and then multi-dimensional feature groups are extracted, knowledge base education contents are matched, and a personalized scheme is generated. And carrying out structural processing on the scheme, constructing a 3D model and a scene resource library, rendering the scheme into a holographic image format, optimizing the scheme, and outputting a material set. And then configuring a calibration material set, scheduling and playing, realizing immersive learning, and generating interaction data and feedback information. And finally, evaluating a learning effect by combining clinical indexes, and dynamically updating and optimizing the knowledge base. By means of 3D holography and multi-mode interaction, intuition, interactivity and individuation of diabetic education are improved, and the self-management ability is enhanced.

The application discloses a guidance text generation method and system based on diabetes management strategy recommendation. The method performs entity recognition on each text to obtain multiple entities, and extracts the relationship between the entities from each text. A knowledge graph is constructed according to the multiple entities and the relationship between the entities. A multi-layer attention layer is used to aggregate the entities and the relationship in the knowledge graph to obtain a first embedding representation. Text embedding of the entities and the relationship in the knowledge graph is generated, and a multi-layer attention layer is used to aggregate the text embedding into a second embedding representation. The first embedding representation and the second embedding representation are aligned and fused to obtain a target embedding representation. According to the target embedding representation and a target problem, multiple management strategy candidate vectors are determined. A target management strategy is selected from the multiple management strategy candidate vectors, and a diabetes management guidance text is generated according to the target management strategy. The application can improve the real-time performance, personalization and intelligentization of diabetes management.

This invention relates to a grading method for diabetic retinopathy based on a local-global interactive dual-branch network, belonging to the fields of deep learning, medical image analysis, and computer-aided diagnosis. It includes: inputting the generated initial feature sequence into a global scanning module; using dilated reparameterized convolution to capture multi-scale structural prior information; inputting the feature map with structural prior information into a core dual-branch LoGo module; the local flow extracting fine-grained lesion features through depthwise separable convolution; the global flow generating spatially variable convolution kernels based on the global context through contextual hybrid dynamic convolution; and performing bidirectional cross-modulation through an adaptive multi-scale feature interaction aggregation module to sharpen local details for global semantics and filter local noise from the global context, ultimately outputting a disease severity grade. This invention explicitly simulates the cognitive strategies of clinicians, significantly improving grading accuracy while maintaining computational efficiency.

A kind of processing method and system of fundus image for diabetes prediction, comprising the following steps: step S100, the original fundus vascular image is preprocessed;Step S200, identify the vascular information in fundus vascular image, and convert to the coordinate and morphology data of fundus vascular and store;It specifically includes: step S210, extract the arteriovenous center line of fundus;Step S220, identify the key point of fundus vascular;Step S230, the identified fundus vascular breakpoint is connected;In the two vascular breakpoints belonging to the same fundus vascular, it is connected using right-angle broken line mode.Step S240, the fundus vascular center line coordinate, fundus vascular bifurcation relationship is stored;Step S300, according to the fundus vascular coordinate and morphology data, extract the fundus vascular feature.This method can also include: step S400, the fundus vascular feature extracted in step S300 is handled by meta-classifier model, to carry out early screening or prediction of diabetes, can greatly improve the prediction accuracy.

The application discloses a kind of based on multi-source medical text construction traditional Chinese medicine diabetic nephropathy knowledge graph method and system, including following module: data integration module: integration diagnosis and treatment guideline and electronic medical record, after redundancy, terminology standardization constructs unified medical data set;Mixed labeling and entity recognition module: fusion artificial labeling and BERT-BiLSTM-CRF model automatically identifies traditional Chinese medical syndrome, traditional Chinese medicine and other complex entities;Relationship modeling module is based on semantic embedding and multi-label classification algorithm, accurately constructs traditional Chinese and western medicine association rules;Atlas application module: entity relationship is converted into attribute graph structure, supports graph database interactive query, realizes diagnosis and treatment recommendation, contraindication early warning and curative effect analysis.Through the above scheme, the present application significantly improves the accuracy of differentiation, reduces the risk of drug contraindication, provides efficient, interpretable decision support for the prevention and treatment of diabetic nephropathy, has important clinical value and popularization potential.

The application discloses a type 2 diabetes heart failure risk prediction system and method, and belongs to the technical field of medical diagnosis and risk assessment. The prediction system comprises the following modules: a data and feature engineering module, which is responsible for standardization processing of data and screening of key prediction factors, and obtains a core feature set for machine learning; a model construction and selection module, which uses the core feature set and trains multiple machine learning algorithms in parallel; through cross-validation and comprehensive performance evaluation, the best model is selected as a prediction model; and a model deployment and application module, which converts output results of the prediction model into a clinically usable static nomogram or online tool, and performs visual output. The application predicts by integrating clinical variables and adopting a machine learning algorithm, and provides a static nomogram and a dynamic Web application, realizes heart failure risk assessment without relying on NT-proBNP detection, and can improve the prevention and management efficiency of cardiovascular diseases.

Disclosed herein is a healthcare management system and method utilizing machine learning based predictive analytics to improve patient adherence rates for chronic care management (e.g., diabetes). An example system may include a computing device configured to establish a customer analytical record to synthetize over a plurality of selected attributes to form a patient centric view of patient behaviors, attitude, characteristics based upon data related to chronic disease state management and social determinants of health, consumer experience to determine, using a predictive artificial intelligence model, a patient's predictive adherence for next 1-3 months, and determine tailored intervention plans to engage patients for continuing adherence.

This invention belongs to the field of diagnostic technology for diabetic complications, and relates to a diagnostic method for type 2 diabetes complications based on multi-source heterogeneous feature transfer. The method includes: extracting features from CGM sequence information, clinical indicators, and past medication information, and dividing them into training and testing sets; using a Gaussian mixture clustering algorithm on the training set to obtain K Gaussian models and the membership class of each sample based on a preset number of clusters K; pre-training samples from each class to obtain K diagnostic models; for each diagnostic model, weighting samples from other classes based on the membership matrix and then transferring the weighted samples from other classes for training to obtain K final models; and using the final models to diagnose complications of type 2 diabetes. Its beneficial effect is that it integrates multi-source information in information processing, solving the problem of fragmentation between sub-models of different subgroups, enabling the models to be fully trained, and improving the accuracy of the models in diagnosing complications.

The application relates to a non-invasive diabetic nephropathy grading and prediction method supporting convenient screening, which comprises the following steps: collecting color fundus images and non-invasive clinical characteristics of diabetic patients, pre-processing to obtain a multi-modal data set, and setting annotations of multiple severity degrees of diabetic nephropathy; constructing a deep learning model A1 to obtain a diabetic nephropathy severity prediction probability distribution based on the color fundus images according to the multi-modal data set; constructing a deep learning model A2 to obtain a diabetic nephropathy severity prediction probability distribution based on clinical data according to the multi-modal data set; constructing a fusion model, the fusion model fuses the obtained diabetic nephropathy prediction probabilities through an adaptive weight distribution algorithm, dynamically adjusts the weights according to the prediction results output by each model, calculates a comprehensive diabetic nephropathy severity score, and outputs a final diabetic nephropathy prediction grading result according to the score.

The invention discloses a guide text generation method and system based on diabetes management strategy recommendation, and the method comprises the steps: carrying out the entity recognition of each text, obtaining a plurality of entities, and extracting the relation between the entities from each text; constructing a knowledge graph according to the relationship between the plurality of entities; aggregating entities and relationships in the knowledge graph by adopting a plurality of attention layers to obtain a first embedded representation; generating text embedding for entities and relationships in the knowledge graph, and aggregating the text embedding into a second embedding representation by adopting a plurality of attention layers; aligning and fusing the first embedded representation and the second embedded representation to obtain a target embedded representation; determining a plurality of management strategy candidate vectors according to the target embedding representation and the target problem; and selecting a target management strategy from the plurality of management strategy candidate vectors, and generating a diabetes management guidance text according to the target management strategy. The real-time performance, individuation and intelligence of diabetes management can be improved.

Disclosed herein is a healthcare management system and method utilizing machine learning based predictive analytics to improve patient adherence rates for chronic care management (e.g., diabetes). An example system may include a computing device configured to establish a customer analytical record to synthetize over a plurality of selected attributes to form a patient centric view of patient behaviors, attitude, characteristics based upon data related to chronic disease state management and social determinants of health, consumer experience to determine, using a predictive artificial intelligence model, a patient's predictive adherence for next 1-3 months, and determine tailored intervention plans to engage patients for continuing adherence.

The invention relates to a diagnosis and treatment data management and collaborative diagnosis and treatment method for diabetic retinopathy, which is applied to a medical care terminal, and comprises the following steps: in response to successful login of a user, displaying a homepage of the medical care terminal; skipping to a medical record auditing page under the condition that the user identity is a medical worker and a request for viewing the medical record auditing page is received; the medical record auditing page is used for displaying the medical record data with the marked auditing state; the medical record data is uploaded by a patient end, detection equipment and / or a hospital information system; when a selection operation on the target medical record data whose auditing state is to be audited is received, displaying the target medical record data in response to a preview operation triggered through a preview component; and when a selection operation on the auditing passing component is received, updating the auditing state of the target medical record data to be auditing passing. Therefore, by means of unified state marking and medical record data summarization, medical record state chaos caused by operation habit differences is reduced, and the problems that medical data are scattered and difficult to integrate are solved.

The invention relates to a data fingerprint method and system for public health data synchronization, and the method comprises the following steps: 1, building a public health service scene data quality control model such as a health record service, a hypertension service and a diabetes service, 2, collecting the existing public health service data, carrying out the ANSIC coding of the data, and storing the data in a database; the method comprises the following steps: 1, carrying out weight statistics on existing basic data, 3, carrying out secondary step-counting classification on the counted data weight, and carrying out recoding arrangement on the data by adopting a binary tree classification method, 4, generating a binary tree number new coding table C1, C2, C3... Cn, and carrying out continuous sorting optimization on the binary tree coding table according to a subsequent data input result, so that the operation efficiency is improved, and the calculation cost is reduced. And 5, generating a public health data field coding file, and carrying out secondary coding to form a smaller field.

The invention provides an auxiliary diabetes diagnosis and treatment system for grassroots doctors, and the system comprises a knowledge graph module which extracts a diabetes prevention and treatment guide and medical terms and entity relationships in an expert consensus unstructured text through natural language processing, constructs a triple knowledge graph containing patient features, clinical guidance and guidance basis, and stores the triple knowledge graph; the input module is used for extracting case texts from the clinical interaction interface, extracting key medical information and examination indexes and sorting the information into a structured format; the logical reasoning module comprises a process controller and an LLM application, the process controller maintains a standardized diagnosis and treatment process and coordinates interaction, and the LLM application receives a structured information generation prompt, calls a large model and generates a visual suggestion; the clinical interaction interface is used for presenting suggestions and receiving doctor input; the feedback self-adaptive module is used for recording operation data, collecting deviation, optimizing a knowledge graph and prompting construction and model parameters; the system can assist primary medical personnel in diagnosis and treatment decision, and improves the diabetes diagnosis and treatment efficiency.

The application discloses a feature selection method for multi-label prediction of diabetic complications, and relates to the technical fields of machine learning and medical data processing. The method comprises the following steps: performing label stratified independent weight calculation on original features and complication label data of the diabetic complications to obtain a plurality of label weight matrices of the complications; performing graph regularization optimization based on the stratified independent weight on the label weight matrices of the complications, training a multi-label classification model, reducing irrelevant and redundant features in the multi-label classification model, and obtaining an important feature subset which is important, has high discriminability and is robust; filtering original data of the diabetic complications by using the optimized multi-label classification model, calculating a feature contribution matrix, and outputting features with top three contribution values for positive prediction labels to construct features for prediction of the diabetic complications. The method can screen out key features and improve the prediction accuracy of the model.

The invention relates to the technical field of medical artificial intelligence and disease risk prediction, and discloses a pancreatic cancer risk prediction model optimization method based on metabolic syndrome data, and the method comprises the steps: a training stage: obtaining historical clinical data of five physiological abnormalities of a sample individual, carrying out the grading assignment of the historical clinical data based on a metabolic syndrome diagnosis standard, and carrying out the prediction of a pancreatic cancer risk prediction model; in the application stage, the clinical data of a target individual are subjected to same preprocessing and feature engineering and input into the risk prediction model, and the risk prediction model is obtained by combining a pancreatic cancer diagnosis tag and training a logistic regression model through a maximum likelihood estimation method. And calculating to obtain a pancreatic cancer risk assessment result of the target individual. According to the method, data modeling is carried out by innovatively integrating metabolic syndrome data and diabetes disease course information, accurate and early individualized prediction of the pancreatic cancer risk is realized, and an effective tool is provided for clinical screening and intervention.

The invention discloses a diabetic nephropathy prognosis method, system and device based on an immunofluorescence image, and relates to the technical field of image feature analysis. The method comprises the following steps: acquiring previous immunofluorescence image data of a patient based on disease detection data disclosed by the patient; determining feature change data in the two images of the patient; and importing the characteristic change data in the two images of the patient into the large system model, outputting the predicted re-visit probability and predicted re-visit time point of the patient, and sending an early warning instruction when the single-day data result of the medical institution exceeds the early warning information threshold value of the medical institution. According to the method, for key links in patient prognosis management, the feature change mode associated with prognosis progress in the sequence immunofluorescence image is systematically extracted and quantified. Based on historical and current iconography data of a patient, probabilistic inference is carried out on a re-doctor-seeing demand in a specific time period in the future and an occurrence time node of the re-doctor-seeing demand, and a quantitative risk assessment result and time interval prediction are output.

The application discloses a multi-modal diabetes dynamic treatment effect evaluation method, relates to the technical field of data processing, and unifies a time axis and marks reliability for multi-source data through an adaptive interpolation and alignment strategy; peak values and abnormal fluctuations within one to two weeks are detected and recognized based on a short-term window, a short-term change amount and a risk prompt are generated; a multi-layer attention or a Transformer is used to fuse data on a monthly scale, comprehensive long-term treatment effect indexes and key inflection points are extracted; short-term quantitative results and long-term indexes are brought into a multi-target reward mechanism through a continuous learning and interactive real-time feedback module, model parameters are iterated online, and clinical intervention suggestions are output; external events are quantitatively marked, and identification and response to sudden situations are strengthened through a bias injection mode. High-quality integration of multi-source heterogeneous data, short-term and long-term multi-level analysis are realized, and the precision and timeliness of diabetes dynamic monitoring and intervention are significantly improved.

The application relates to the technical field of medical data processing, and discloses a diabetic nephropathy risk assessment model. The model comprises collecting time-series medical data of a diabetic patient, including blood test values, body measurement data and diagnosis and treatment records. For each time point, feature elements are extracted and a temporary risk assessment value is calculated. A historical medical database is connected, similar patient categories are matched, the temporary risk assessment value is corrected by using the risk history of the similar patient categories, and a unified risk assessment value is obtained. The difference between the temporary risk assessment value and the unified risk assessment value is taken as an accuracy index. In combination with the accuracy index and the time-series change of the features, the future nephropathy risk grade of the patient is predicted. The application calibrates individual assessment by using group historical data, generates an internal accuracy index, and improves the accuracy of risk assessment and the reliability of clinical decision-making.

The invention discloses a knowledge graph large model extraction method and system based on modeling cue words, and belongs to the technical field of natural language processing. Entity extraction and relation extraction task signatures special for the diabetes field are defined, and input and output formats and type constraints are clarified; constructing a hierarchical strategy optimizer, learning an optimal cue word template and an example selection strategy through a training process, and converting a declarative task signature into efficient cue word engineering implementation; loading the trained optimizer model, and guiding the large language model to perform gradual reasoning through a reasoning chain mechanism, so that the large model performs gradual reasoning according to a sequence of firstly extracting entities and extracting a relationship based on the entities to form a structured knowledge extraction process; establishing a multi-dimensional evaluation mechanism, and evaluating an extraction result; high-precision and high-efficiency automatic construction of the diabetes knowledge graph is realized under a small amount of labeled data, and the method is suitable for intelligent diagnosis, treatment scheme recommendation and medical research support in the field of medical health.

The present invention relates to a system for providing diabetes educational content customized on the basis of health literacy, and a method therefor. According to the present invention, the system may comprise: a display unit for displaying, to a user, a plurality of pre-stored health literacy evaluation questions or diabetes educational content; an input unit for receiving an response of the user for each displayed health literacy evaluation question; an analysis unit for evaluating health literacy of the user by assigning, to the input responses of the user, a pre-set score for each question; and a control unit for transmitting a control signal so as to display the pre-stored diabetes education content on the display unit according to the health literacy of the user.

The invention is an AI-driven health screening system for early detection of cancers, cardiovascular diseases, and type 2 diabetes. It integrates multiple screening methods, using machine learning, deep learning, and natural language processing to analyze genetic and health data. A federated learning framework ensures high prediction accuracy while maintaining data privacy. The system also employs a blockchain-based electronic health record (EHR) for secure data management. AI models, including neural networks and support vector machines, assess risk factors and provide personalized healthcare recommendations. Designed with a three-tier architecture, it supports deployment as a web service, software, or integration into existing programs. The system enhances early disease detection, optimizes healthcare resources, and improves patient outcomes by offering a scalable, efficient, and non-invasive screening solution.

The invention discloses a diabetes risk prediction method based on high-dimensional disease embedding and health data fusion. Comprising the steps of data acquisition and preprocessing, construction of a high-dimensional disease embedding space, calculation of an individual health condition vector, deduction of a health score vector based on physical examination data, a hybrid prediction model for risk assessment, and model training and deployment. According to the method, a hybrid machine learning architecture is adopted, firstly, a health condition vector, a health score vector and original physical examination data are effectively integrated through a special fusion module, and then fused features are input into an integrated prediction model to generate a highly accurate diabetes risk score; according to the method, the prediction performance is remarkably improved, and the area (AUC) under the working characteristic curve of a subject reaches 0.84.

According to the diabetic retinopathy intelligent grading method based on enhanced OfficientNetB0 and MobileNetV1 fusion provided by the invention, a new fusion model is designed and realized after image preprocessing and image expansion, multi-scale feature complementation is realized by improving an OfficientNetB0 model and fusing a MobileNetV1 model, and the grading accuracy is remarkably improved. The improvement of the OfficientNetB0 model comprises the steps of introducing an RLGA module behind a top layer feature map to enhance the feature extraction capability, and realizing the refinement of local features and the enhancement of global feature perception by combining a local perception module (LPM) and a global local attention module (GLAM). Meanwhile, the classification performance is improved by utilizing the offset NetB0 transfer learning, 3 * 3 deep convolution is added in a residual shortcut branch, gradient transfer is avoided, learning features are reinforced, so that the evaluation index is further improved, finally, the improved offset NetB0 and the MobileNetV1 are fused, and the classification precision of the model is improved by combining the advantages of a composite scaling strategy of the offset NetB0 and the depth separable convolution of the MobileNetV1.

A method for generating synthetic data is provided, comprising receiving, via an input device, input data comprising input first data indicative of a first healthcare-relevant parameter, the input first data providing at least one sparse time series input signal for the first healthcare-relevant parameter, and determining, in a processing device, auxiliary first data from the input first data. The sparse time series input signal of the input first data is encoded into a dense time series input signal for the first healthcare-related parameter, the auxiliary first data providing the dense time series input signal. In the processing device, synthetic data is determined from modified input data comprising the auxiliary first data such that the synthetic data closely resembles the modified input data. Further, a corresponding synthetic data generation device, a computer program product, a method for predicting a healthcare-relevant parameter, and a diabetes management device are provided.

The present invention relates to a method for predicting the prognosis of intravitreal injection treatment, comprising the steps of: receiving OCT images and clinical information of a plurality of diabetic macular edema patients; extracting one or more major lesion information from the OCT images; training a multi-modal deep learning model by inputting the OCT images, the clinical information, and the one or more major lesion features, and outputting the intravitreal injection treatment prognosis of the diabetic macular edema patients; and predicting the prognosis of intravitreal injection treatment by inputting the OCT images, the clinical information, and the major lesion features of a specific diabetic macular edema patient to the trained multi-modal deep learning model.

The system of the embodiment aims to personalize and optimize diabetes risk assessment and management. [Solution] A system according to an embodiment includes a collection unit, a prediction unit, a provision unit, a monitoring unit, and a platform unit. The collection unit collects health checkup data. The prediction unit analyzes the data collected by the collection unit to predict a diabetes risk score. The provision unit provides individual risk assessment and advice based on the risk score predicted by the prediction unit. The monitoring unit collects and analyzes data using monitoring devices. The platform unit provides a patient community platform.

This invention provides a computer-aided analysis system based on high-throughput sequencing data and knowledge graphs, including a data access agent, a quality control analysis agent, a species annotation agent, a knowledge graph reasoning agent, a report generation agent, and an agent scheduling center. The data access agent acquires sequencing data and clinical phenotypic data; the quality control analysis agent performs quality control using the Illumina NovaSeq platform, V3-V4 primers, the SILVA v144 database, Q30 ≥ 90%, and sequences ≥ 50 bp; the species annotation agent performs species classification annotation based on a reference database; the knowledge graph reasoning agent performs association analysis based on a microbiome-disease-intervention knowledge graph, using ET_MGNN and RTGN models, including 264 search keywords covering dental caries, periodontitis, oral cancer, Alzheimer's disease, diabetes, and heart disease; the report generation agent generates personalized analysis reports and intervention suggestions; and the agent scheduling center manages and dynamically schedules all agents, forming a complete closed loop of "perception → analysis → diagnosis → intervention → execution → feedback".