1423results about "Laboratory analysis data" patented technology

The invention provides a multi-modal large language model for generating a hepatocellular carcinoma key pathological diagnosis report, a framework main body is a visual coding module, and a multi-modal feature alignment module, a multi-head low-rank attention mechanism, an enhanced medical MoE mechanism and a structured output decoding layer are also introduced. The visual coding module is constructed on the basis of a Swin Transform architecture, visual pre-training is completed on hepatocellular carcinoma MRI data, and after a task specific classification head is stripped, a trunk feature extraction network is reserved to serve as an image modal representation encoder. The multi-modal feature alignment module guides the model to learn a cross-modal semantic mapping relation between a hepatocellular carcinoma MRI image and a key pathological diagnosis report language, image modal input is a visual feature sequence, and text modal output is a structured description text; and the structured output decoding layer generates six types of liver cancer focus attributes. According to the method, the pre-operative multi-parameter and multi-stage enhanced MRI image is utilized, and the open-source large model is finely adjusted to generate a matched liver cancer postoperative pathology report.

The invention provides an intelligent medical information system based on multi-system integration, a data security management method and device, electronic equipment and a storage medium, and is applied to the technical field of medical data processing. According to the method, three types of information including core security elements, multi-system heterogeneous adaptation (including HIS / LIS / EMR interface specifications and the like) and full-process auditing (including operation logs and the like) are obtained. Normalizing the core security elements by using technologies such as AI multi-mode privacy identification and the like to generate a standardized security baseline file; heterogeneous adaptation information is processed through a lightweight API connector and the like, and a multi-system security access scheme is formed. Associating the baseline file with the access scheme, and generating a full-life-cycle protection strategy in a zero-trust architecture; and processing the audit information to generate a dynamic security audit report. And finally, based on a security-adaptation linkage decision engine and compliance verification, outputting a cross-system data security sharing scheme and a privacy disclosure emergency response instruction, and realizing a medical data security management closed loop.

The invention discloses a method and system for predicting juvenile depression based on intestinal flora, and relates to the technical field of bioinformatics and artificial intelligence, and the method comprises the steps: firstly, obtaining an original sequence of a microbiome, carrying out the preprocessing of the original sequence of the microbiome, and obtaining a feature matrix; and screening core flora characteristics with stable trans-folding by adopting characteristic importance evaluation and interpretability analysis based on a gradient boosting decision tree. A mixed weighted graph is constructed based on Spearman correlation and a proximity relationship, and an absolute value of a correlation coefficient is taken as an edge weight and an edge density is adjusted through a threshold adaptive strategy. And finally, through an improved graph attention neural network, based on edge weight attention, layer normalization and random inactivation, enhancing robustness, and adopting adaptive optimization to complete parameter learning. And determining a dynamic classification threshold according to the AUC of the target patient, and outputting a sample discrimination result and confidence. According to the method, the accuracy, stability and biological interpretability of juvenile depression recognition are remarkably improved.

Methods and systems for performing a clinical prediction are provided. In one example, the method comprises: receiving first molecular data of a patient, the first molecular data including at least gene expressions of the patient; receiving first biopsy image data of the patient; processing, using a machine learning model, the first molecular data and the first biopsy image data to perform a clinical prediction of the patient's response to a treatment, wherein the machine learning model is generated or updated based on second molecular data including at least gene expressions and second biopsy image data of a plurality of patients; and generating an output of the clinical prediction.

The invention relates to the technical field of artificial intelligence, in particular to a machine learning-based double-J tube indwelling membranous fungus infection risk medication decision support method, and aims to solve the problems of inaccurate double-J tube indwelling membranous fungus infection risk assessment and medication decision experience. According to the method, the machine learning model for predicting infection risks, pathogenic bacteria and drug curative effects is constructed by collecting multi-dimensional data and performing feature engineering, and personalized medication suggestions and visual decision support are provided in combination with a knowledge base and a rule engine. By adopting the scheme, accurate risk prediction can be realized, drug curative effect evaluation can be optimized, drug use reasonability and safety can be ensured, and clinical decision-making efficiency can be improved.

The invention discloses a medical knowledge graph construction method and device for multi-source heterogeneous data fusion and incremental updating, and relates to the technical field of medical knowledge graphs. The method comprises the steps that multi-source medical data such as outpatient records, hospitalization medical records, inspection reports and image data of a patient in the whole life cycle are collected; performing time sequence alignment by adopting a space-time alignment algorithm to obtain standardized medical data, calculating a data fusion weight and constructing an incidence matrix; performing singular value decomposition on the incidence matrix to extract a feature vector, and obtaining a dynamic update coefficient through standardization, cosine similarity calculation, convolutional neural network processing and exponential weighted moving average; adjusting the reference map according to the dynamic updating coefficient increment to generate an updated medical knowledge map, and outputting diagnosis and treatment decision support information; and multi-dimensional medical data space monitoring abnormity can be constructed, a joint optimization model is established to generate a personalized diagnosis and treatment scheme, and edge deployment is realized through a knowledge distillation compression map.

The invention relates to the technical field of medical treatment, and particularly discloses a compound prognosis model for cancer prognosis evaluation, a prognosis evaluation method and a calculator. The compound prognosis model is constructed based on inflammation burden related prognosis indexes ALNCR and TNM by stage combination; the creation method comprises the following steps: step 1, collecting albumin concentration, lymphocyte count, neutrophil count and C-reactive protein concentration data of a cancer patient; 2, calculating an ALNCR index according to a formula; 3, according to a clinical classic TNM staging system, the patients are divided into a TNM staging stage I, a TNM staging stage II, a TNM staging stage III or a TNM staging stage IV; 4, the ALNCR index obtained in the step 2 and the TNM staging result obtained in the step 3 are associated and integrated, and the individualized prognosis score of the ALNCR.TNM composite prognosis model is obtained through calculation. The method is suitable for clinical rapid application; high-risk patients can be accurately distinguished, a quantitative basis is provided for individualized treatment and follow-up visit strategy formulation, and excessive medical treatment or insufficient treatment is reduced.

A device may include a data acquisition layer comprising a patient information collector and a drug information collector, each operable to ingest patient-specific clinical data and drug-specific data, respectively. A device may include an analysis engine comprising a multi-criteria ranking engine configured to compute composite suitability score for candidate medications by weighting one or more of medication efficacy, medication safety, medication resistance, medication cost, or patient-specific coverage factors. A device may include a presentation layer comprising a summary dashboard configured to display a ranked medication recommendation set produced by the analysis engine.

A Health Value Analytics (HVA) system includes an electronic medical record (EMR) / HVA access device in the form of comprising machine instructions for facilitating a physician's workflow. A processor at a physician's work terminal executes the machine instructions to connect the terminal with a HVA server and / or an EMR server. The processor sends a medical information request for a patient and information related to characteristics and parameters of the terminal, to one or both the HVA server and the EMR server. The processor receives, in response, an EMR page generated and adapted by the EMR server for display on the terminal. The EMR page displays an electronic medical record of the patient, and one or more HVA data objects generated by the HVA server, after the EMR server adapts the EMR page based on mode information that includes characteristics of the terminal.

Presented herein are systems and methods for classifying biomedical images for executing operations. A computing system may identify a biomedical image of a slide with a biological sample obtained from a subject at risk of a condition; apply the biomedical image to a machine learning model; generate, based on applying the biomedical image to the ML model, a classification corresponding to the biomarker associated with the condition in biological sample on the slide; and execute an operation with respect to the slide for testing of the biological sample, in accordance with the classification for the biomedical image.

The invention relates to the technical field of data management and information systems, in particular to an experimental data management system conforming to GLP specifications, which comprises a data acquisition and standardization unit, a data storage and version management unit, a natural language query and interaction unit, a data quality and compliance verification unit and an audit tracking and authority management unit. According to the method, professional terms and query intentions are accurately analyzed by means of a field-specific language model, intelligent recognition and visual interaction of image data are realized by the multi-modal interaction module, dialogue continuity is maintained by the context dialogue management module, and an association analysis scheme is actively pushed by the intelligent recommendation module. Intelligent retrieval and analysis requirements of animal experiment data are efficiently supported, the compliance and quality of the experiment data are guaranteed, and data risks are effectively identified and handled.

The invention provides systems and methods for diagnosing and treating an inflammatory skin disease in a patient. The described methods may be used for longitudinal monitoring of the development, progression, or regression of a skin disease including lupus, psoriasis, atopic dermatitis, and systemic sclerosis (scleroderma) condition in a patient, in response to a therapy administered for the condition.

Provided herein are computerized system and method for full patient flow management through an emergency department process, integrating multi-agents LLM. The system and method utilize a plurality of LLM-based agents configured to perform one or more of: extract relevant medical information and generate a summarized medical history; interactively questioning the subject regarding its medical history and regarding the main complaint(s); provide recommendations regarding various medical examinations; process the summarized medical history and the results of the recommended medical examination(s) and; provide further summary, recommendations or instructions, regarding diagnosis and a decision regarding discharge or admission of the subject.

A method for automatically generating a research and development proposal for an in vitro diagnostic kit, performed by a server for automatically generating the research and development proposal, includes: acquiring an initial proposal including intended use information for the kit; providing the acquired initial proposal to a pre-trained language model; acquiring a specification (spec) for the kit from the language model receiving the initial proposal, wherein the acquired specification for the kit includes at least one of a technical specification to be used for the research and development of the kit, a resource specification for the research and development, a design specification for the kit, a regulatory approval specification that is needed to meet for approval, and a performance specification for the kit; and controlling to generate the research and development proposal for the kit including the acquired specification and the intended use information for the kit.

Provided is an application of a circular RNA in the preparation of a cerebral stroke diagnosis product, belonging to the field of circular RNA medical treatments. Provided is a kit, comprising a reagent and / or test paper and / or gene chip for detecting a circular RNA-circ-Magi1. The nucleotide sequence of the circular RNA-circ-Magi1 is as shown in SEQ ID NO: 1 or SEQ ID NO: 4. Further provided is the application of a detection system for the diagnosis, assisted diagnosis or prognosis evaluation of cerebral stroke, and a detection reagent of the circular RNA-circ-Magi1 in the preparation of a kit, test paper, chip or high-throughput sequencing platform for the diagnosis, assisted diagnosis or prognosis evaluation of cerebral stroke.

The invention discloses a medical informatization-oriented test sheet structured identification and automatic input method, and relates to the technical field of medical informatization. The method comprises seven steps of inspection single-mode and multi-mode image acquisition, intelligent image preprocessing and optimization, medical special OCR joint recognition, structural data intelligent analysis, cross-system data verification and matching, hierarchical data automatic input and full-process tracing and manual rechecking. The image quality is guaranteed through a customized image acquisition and preprocessing technology, the complex content recognition accuracy is improved by using a medical special OCR model, data structuring and accurate matching are realized by relying on a multi-stage analysis rule base and a cross-system verification mechanism, and the data reliability is ensured by combining hierarchical entry and manual recheck. And meanwhile, full-process tracing and system flexible adaptation are supported. According to the invention, the problems of low processing efficiency, poor accuracy, insufficient adaptability and the like of the inspection sheet in the prior art are solved, rapid, accurate and automatic input of the inspection sheet information is realized, medical informatization construction is effectively supported, and the medical service quality and efficiency are improved.

A method, a system, and computer program product for managing tokenized biospecimen transfers are provided. A file including private data associated with one or more donors is retrieved by a biobank node connected to a blockchain network and from a database of a data provider. The file is processed to generate tokens corresponding to the private data, each token representing a biospecimen and being assigned to a respective donor. An inventory is updated by adding identifiers of the plurality of tokens to the inventory. A request to transfer one or more tokens corresponding to a portion of the private data is received. An authorization to transfer the one or more tokens to a target entity is received. The token is transferred to the target entity with a consent to perform one or more operations using the token.

In some aspects, a triage artificial intelligence receives a portion of a conversation between a doctor and a patient, determines, based on the conversation, one or more symptoms associated with the patient, determines that the conversation includes a set of trigger words associated with a particular symptom of the one or more symptoms, selects based on the set of trigger words associated with the particular symptom, a first artificial intelligence specialist in a first medical specialty from a plurality of artificial intelligence specialists, and provides the first artificial intelligence specialist with access to the conversation between the doctor and the patient. The first artificial intelligence specialist provides one or more decision support insights that are sent to a computing device associated with the doctor.

The invention provides a biological sample scheduling method and device, a storage medium and a program product. The method comprises the following steps: acquiring sample information of at least one scheduling unit; screening the sample information of the scheduling unit according to a preset priority standard to obtain a plurality of priority samples; the biological samples of the remaining scheduling units are subjected to order splitting and order splitting to obtain a plurality of second aggregation units, the order splitting is used for combining the biological samples of at least one remaining scheduling unit after screening according to a preset priority standard, and the order splitting is used for splitting the biological samples of the remaining scheduling units after order splitting; and generating a scheduling list according to the sample information corresponding to the first aggregation unit and the second aggregation unit, the position layout of the biological sample of each first aggregation unit and the biological sample of each second aggregation unit on the pore plate, and the distribution information of preset priority standard screening, order halting and order splitting. The problem of low distribution efficiency of biological samples in the prior art is solved.

The invention discloses a multi-factor risk assessment method for ovarian hyperstimulation, and relates to the technical field of intelligent medical treatment, and the method comprises the steps: based on a multi-source baseline data packet, calculating the trend of follicle volume along with time through ovarian reaction simulation to obtain a follicle reaction curve, calculating the probability of vascular injury through puncture path bleeding simulation, and calculating the risk of ovarian hyperstimulation according to the probability of vascular injury. Obtaining a bleeding probability, and fusing the follicle reaction curve and the bleeding probability to form a patient characterization risk vector; associating the intervention suggestions in the structured strategy list with the ovarian overstimulation risk score and the bleeding risk score to generate a risk change description, and forming a risk assessment report according to the risk change description; according to the method, the patient characterization risk vector fusing the follicle dynamic response and the puncture bleeding risk is constructed, so that the safety and individuation level of assisted reproduction treatment are effectively enhanced.

The invention discloses an early pancreatic cancer prediction and risk stratification system based on artificial intelligence, and belongs to the technical field of medical health data analysis and artificial intelligence. The system comprises a multi-omics data adaptive fusion module, a longitudinal health trajectory coding module, a biomarker combination discovery module, a risk prediction and dynamic layering module and a closed-loop feedback optimization module, and a data confidence index generated by the multi-omics fusion module directly affects the attention weight of longitudinal trajectory coding. Longitudinal track coding adopts a bidirectional long-short-term memory network to extract time sequence characteristics, a biomarker discovery module recognizes a synergistic marker combination through a Transform mechanism, a closed-loop feedback module dynamically adjusts parameters of each module according to a prediction result, and clinical verification shows that the prediction accuracy of the system reaches 85%, the I-stage diagnosis rate is improved by 60%, diseases are discovered 8-12 months in advance, and the diagnosis efficiency is improved. The method is obviously superior to the prior art.

Described herein are methods such as multi-omic methods for assessing a disease such as cancer. The multi-omic methods may integrate proteomic, transcriptomic, genomic, lipidomic, or metabolomic data. The method screening diseases or disease states. Also described herein are methods for screening for diseases or disease states from biological samples. The methods may include assessing whether a nodule, mass, or cyst is cancerous.

The invention discloses a training method, system and device of a disease auxiliary diagnosis model and a medium. The method comprises the steps of preprocessing local multi-modal medical data of each mechanism; performing adversarial training on the multi-modal medical data through a generator and a discriminator to generate a high-fidelity synthetic sample; updating gradient parameters of the disease auxiliary diagnosis model based on the synthetic sample; the gradient parameters are encrypted and uploaded to a federal center server; the server aggregates the gradients and issues the gradients to each mechanism; and generating a disease auxiliary diagnosis model after iterative optimization. According to the method, scarce samples are supplemented through multi-modal data standardization and adversarial generation, cross-institution privacy protection cooperation is realized in combination with federated learning and a differential privacy technology, a global optimal model can be constructed without sharing original data, the model diagnosis accuracy and generalization ability are improved, the problem of data imbalance is effectively relieved, and the user experience is improved. And clinical auxiliary diagnosis and medical scientific research requirements are met.

The invention discloses an ovarian cancer postoperative low blood volume shock risk dynamic prediction system, and relates to the field of clinical prediction models, and the system comprises a clinical parameter collection module which is used for obtaining a plurality of perioperative period clinical parameters related to postoperative low blood volume shock; the risk calculation engine performs weighted integration on the clinical parameters based on a multi-factor regression model to generate an individualized risk score; the visual output module is used for displaying the contribution degree of each parameter and the overall risk level; and the real-time risk trend prediction module is used for collecting physiological parameter dynamic change data at multiple time points after an operation, quantifying a risk change trend by calculating a risk evolution index, and fusing the index with a static risk value to generate a final risk score. According to the scheme, the early recognition capability and prediction precision of the postoperative low blood volume shock risk can be remarkably improved.

A computer-implemented method of training a machine learning model to predict a clinical outcome or characteristic based on a patient's clinical history is disclosed. The method comprises: providing training data comprising structured electronic health record data for a plurality of patients, the structured electronic health record data comprising a plurality of clinical observations, each clinical observation having a text description and an associated time stamp, wherein the training data for each patient is labelled with one or more labels, each representing a clinical outcome or characteristic; converting each patient's electronic health record data into a text sequence comprising the text descriptions concatenated in sequence of the time stamps; inputting the text sequence into a machine learning model; and training the machine learning model to predict a clinical outcome or characteristic based on the input text sequence.

The present disclosure describes methods and systems for predicting if a subject has an increased risk of having or developing one or more clinical outcomes, including prior to the detection of symptoms thereof and / or prior to onset of any detectable symptoms thereof. The present disclosure also describes a method of generating a model for predicting one or more clinical outcomes.

The invention relates to the technical field of maternal and fetal medicine, in particular to a regional maternal and fetal medicine congenital structure anomaly risk assessment system based on big data. A regional feature quantification module; a risk assessment model module; an interpretability analysis module; and a streaming update module. According to the scheme, deep integration and unified representation of individual information and regional macroscopic attributes of pregnant and lying-in women are realized by constructing a multi-source heterogeneous data fusion and regional feature quantification mechanism. The system adopts a multi-protocol interface to access a hospital information system, an environment monitoring platform and a health statistics database, completes standardized cleaning and integrated processing of cross-domain data, and converts multi-dimensional indexes such as regional environment, medical resources, social economy and the like into low-dimensional dense feature vectors through a macroscopic data aggregation and ridge regression coding technology. The feature expression ability and the regional adaptability of the model are significantly enhanced, and effective data are provided for evaluating the risk of the innate structure anomaly of pregnant and lying-in women under different geographical backgrounds.

Provided is a method for assisting in grasping the status of platelet activation, by developing a biomarker that more accurately reflects platelet activation in vivo, said method being usable in testing for thrombotic hemostatic diseases. The method for assisting in grasping the status of platelet activation comprises (1) measuring the concentration of soluble CLEC-2 and a platelet count in a sample obtained from a subject, and (2) calculating a value of [soluble CLEC-2 concentration] / [platelet count] by dividing the soluble CLEC-2 concentration by the platelet count. If desired, the method may further comprise (3) comparing the value with a value of [soluble CLEC-2 concentration] / [platelet count] obtained by using samples from healthy persons.

One or more instruments generate test result data. The test result data include patient data and QC data. The test result data is provided to a QC data flow system via a local network, which filters the test result data (e.g., using a set of rules) to extract the QC data. The QC data is provided to a cloud-based QC data management platform via an external network. The cloud-based QC data management platform analyzes the QC data and provides a result back to the QC data flow system. The QC data flow system forwards the result to middleware or the instrument, which triggers a corrective action based on the result as appropriate.WO

This provides an evaluation method for constructing robust quantitative indicators from tissue images. [Solution] The evaluation method provides cell nucleus contour data to each cell in the tissue image, assigns a first boundary line to one side of the tissue image and a second boundary line to the other side, and places n (n: an integer greater than or equal to 1) isometric lines in the region of the tissue image sandwiched between the first and second boundary lines to partition (n+1) layers. Here, an isometric line is a line consisting of a set of points where the ratio of the distance to the first boundary line to the distance to the second boundary line is equal, and the ratio of the distance to the first boundary line to the distance to the second boundary line at the k-th isometric line (k is an integer greater than or equal to 1 or less than or equal to n) from the first boundary line satisfies k:n+1-k. The evaluation method further determines a first index based on the contour data, sets a region of interest consisting of one or more layers, and determines a first statistical value for the first index in the region of interest.