420results about "Pathological references" patented technology

The invention relates to the technical field of medical big data analysis and intelligent monitoring, in particular to a medical time series data anomaly detection system which comprises a multi-modal data fusion module used for obtaining a physiological sensor data stream of a target object, performing multi-source heterogeneous synchronization and tensor coding on the physiological sensor data stream, and obtaining a multi-modal data fusion result; constructing a multi-modal physiological time sequence tensor; and the phase-space reconstruction module is used for performing high-dimensional dynamic mapping on the multi-modal physiological time sequence tensor. The one-dimensional time sequence signals are mapped to the high-dimensional Euclidean space through the phase-space reconstruction module, the dynamic manifold structure of the physiological system is restored, abnormity is recognized by detecting the morphological variation of attractor tracks in the high-dimensional space, and even if the physiological parameters do not reach the alarm threshold value in numerical value, the abnormity is recognized. As long as an internal nonlinear dynamic structure is changed, the system can carry out sensitive capture, so that the problem that a traditional system misses detection of early-stage hidden pathological features is effectively solved.

The invention relates to a pathological section intelligent auxiliary differential diagnosis system based on multi-modal fusion, in particular to the field of clinical pathology, semantic unification of multi-modal data is achieved through meta-task construction and a cross-modal alignment technology, and transferable diagnostic knowledge is extracted by utilizing a meta-learning framework; the method combines a generative model and knowledge constraints to generate high-quality synthetic data, and finally fuses real and synthetic samples through a self-adaptive diagnosis mechanism, thereby remarkably improving the differential diagnosis capability of rare lesions, effectively solving the problem of model generalization in a training data scarcity scene, and improving the accuracy of model identification. And efficient and reliable intelligent auxiliary decision support is provided for clinical pathological diagnosis.

The invention discloses a hepatocellular carcinoma postoperative early recurrence prediction method based on multi-modal fusion. The method comprises the following steps: firstly, integrating clinical data of a training set, a preoperative enhanced CT image and a postoperative full-view digital pathological image, and carrying out standardized correction; then, traditional image omics features and deep learning features are extracted from the CT image, cell nucleus morphological features and tumor microenvironment spatial configuration features are extracted from the pathological image, and key feature signatures are screened out through a maximum correlation minimum redundancy algorithm (mRMR) and LASSO regression in combination with clinical features. And then carrying out progressive model construction by adopting an XGBoost algorithm, sequentially establishing a clinical single-mode model, an image single-mode model, a pathological single-mode model and a multi-mode fusion model, and explaining and visualizing the models by utilizing an SHAP value and a Grad-CAM technology. Finally, the performance of the model is evaluated in a multi-dimensional mode through internal cross validation, foresight and external independent validation, risk layering is carried out based on the prediction probability, and individualized postoperative management is guided.

The invention discloses a personalized health management method and system based on an AI electronic medical record, and relates to the technical field of artificial intelligence medical treatment, and the method comprises the steps: analyzing an original electronic medical record to generate a personal health timeline; carrying out feature extraction on the health state evolution sequence, identifying key nodes, and carrying out pathological labeling according to a medical knowledge graph to form a health state evolution sequence with a pathological label; a risk assessment model is constructed, future disease risks are calculated based on the sequence, and a dynamic report is generated; making a personalized health management plan in combination with the living habits and genetic backgrounds of the users; during plan execution, user feedback and monitoring data are collected in real time, and plan content and strength are dynamically adjusted by using a reinforcement learning mechanism. According to the method, the medical interpretability of health state evolution is enhanced through pathological labeling, and dynamic closed-loop optimization of a management plan is realized through reinforcement learning.

The invention discloses a glioma radiotherapy postoperative risk assessment method based on a magnetic resonance image, particularly relates to the field of glioma radiotherapy patient health risk assessment, and is used for solving the problem that an existing assessment mode depends on manual interpretation and is difficult to predict bad clinical outcomes in advance. The method comprises the following steps: performing clinical data gridding reconstruction on a corticoid use cycle of a patient and tumor molecular typing, and combining morphological characteristics of an edema region in a magnetic resonance image to generate time-aligned clinical comprehensive characteristic vectors; mining a frequent association item set between the comprehensive feature vector and the pathological process to construct a mapping relation model; establishing a probability graph reasoning model of the bad outcome based on the pathological process vector sequence and the probability weight; and integrating the two types of models to form a causal reasoning network, and inputting a target patient feature vector to calculate an accumulated risk value of reaching a bad outcome. According to the method, an interpretable individual risk assessment result can be output, and a basis is provided for postoperative follow-up visit and intervention.

The invention discloses a newborn health monitoring system based on multi-modal data fusion, and belongs to the technical field of medical monitoring. According to the system, aiming at the problem of misjudgment caused by complex background interference and newborn development difference, a background perception feature decoupling enhancement network is constructed, a dual-channel architecture is matched with an adversarial loss function, foreground skin and background features are forcibly separated, and weak pathological features are enhanced in combination with a color space attention mechanism. Meanwhile, a development stage self-adaptive classification network is utilized, the gestational age serves as prior information to generate development codes, and dynamic modulation and weighted evaluation are conducted on fusion features through a premature infant and full-term infant double-branch classifier. According to the invention, the influence of environmental noise is effectively eliminated, and differentiated accurate health state recognition and risk early warning are realized.

Provided is a method for screening myocardial therapeutic targets for ischemic heart failure by using single-cell sequencing, which method comprises the following steps: S1, sample preparation; S2, construction of a single-cell expression matrix; S3, cell quality control; S4, cell type annotation; S5, cell communication analysis; and S6, co-expression network analysis. The provided method for screening myocardial therapeutic targets for ischemic heart failure by using single-cell sequencing comprises performing single-cell sequencing on hearts of healthy mice and IHF mice, screening for cell types with significant differences in cardiac transcriptional profiles of the healthy mice and IHF mice, then exploring interaction characteristics of various types of cells in malignant fibrotic IHF hearts, revealing potential regulatory modules and pathways related to malignant myocardial fibrosis in single-cell expression data of IHF hearts, and performing screening to obtain Pdgfb and Tnfsf12 genes which can be used as therapeutic targets for treating myocardial fibrosis in ischemic heart failure.

The invention provides a pathological section human-like reading track generation method based on reinforcement learning. The method comprises the following steps: constructing a training data set; the training data set comprises a plurality of WSIs and corresponding doctor film reading track data; an RL frame is built, and parameters of the built RL frame are initialized; wSI local image features and a WSI current film reading state are taken as a state S, position movement in eight directions and a preset fixed step length is taken as an action A, and a pathological expectation value output by a PEAN model is taken as a reward R; and training a PEAN model agent based on the deep reinforcement learning Q network and a sequence of the state S, the action A, the reward R and the next state S stored in the experience playback pool to realize iterative optimization of the deep reinforcement learning Q network so as to finally generate a human-like film reading track of which the coincidence degree with the doctor film reading track is greater than or equal to a preset coincidence degree. According to the method, the macroscopic and microcosmic film reading logic of a doctor is reproduced, and the WSI diagnosis efficiency is greatly improved.

The invention relates to the technical field of medical data analysis, in particular to a urinary system tumor big data analysis system which comprises a target kernel generation module, a kernel matrix calculation module, a parameter optimization module and a risk division module. According to the method, a target kernel matrix reflecting clinical prognosis differences is constructed and serves as an optimization reference, a radiomics and genomics feature kernel matrix is generated through hardware acceleration parallel computing, kernel function width parameters are dynamically iteratively updated based on an alignment degree numerical value so as to ensure that multi-modal feature distribution is highly matched with a prognosis label, and the accuracy of the multi-modal feature distribution is improved. A multi-dimensional feature space containing rich pathological information is constructed by combining a weighted fusion mechanism after centralization processing, so that a support vector machine is trained to determine a high-robustness decision boundary, and precise division of tumor risk levels is realized while high-dimensional data calculation delay is greatly reduced; and the reliability and timeliness of auxiliary diagnosis and treatment results in a complex pathological environment are effectively improved.

The application relates to the technical field of data processing, and discloses an old person health state evaluation data processing system based on multi-modal data, which comprises a medical data integration module, an across-modal causal fusion processing module, a health state evolution modeling module, a backtracking evaluation report module and a decision output module. The medical data integration module acquires electronic medical record data and medical examination reports through a FHIR interface and extracts structured health indexes. The across-modal causal fusion processing module fuses monitoring data and medical texts through an image, a text and an image sandwich architecture. The health state evolution modeling module maps a health feature vector into three-dimensional state indexes of physiological functions, cognitive levels and motor abilities. The backtracking evaluation report module and the decision output module are used for backtracking evaluation and decision output. Through the image, the text and the image sandwich architecture, deep semantic fusion of multi-modal features is realized under the constraint of medical pathology rules, feature weight adaptive distribution is dynamically guided based on a medical causal graph, a high-dimensional fusion vector retaining key pathological information is generated, the semantic integration capability of health data is significantly improved, and the reliability of discrimination and decision is improved.

Embodiments of the disclosure include systems and non-transitory computer readable media for analyzing microscopy images for developing machine learning models for disease modeling. Microscopy images are captured from cells of one or more exposure response phenotypes (ERPs) and further used to train machine learning models. Thus, trained machine learning models can distinguish between microscopy images captured from healthy and diseased samples.

An ideation platform device and method using a diagram are disclosed. An ideation platform device using a diagram, according to one embodiment of the present invention, can comprise: a C-K canvas module for providing a C-K canvas divided into a concept space and a knowledge space and connecting a concept and knowledge to each other on the C-K canvas through a chaining process so as to help a solution search for resolving a problem; and an instance management module for storing and managing, as one instance, the C-K canvas, for which a solution search is completed, including the concept, the knowledge, and information about an interconnection relationship.

The invention relates to the technical field of adaptive control, in particular to a syndrome differentiation adaptive control system for intelligent traditional Chinese medicine treatment equipment, which comprises an environmental parameter acquisition unit for acquiring microenvironment parameters in real time and generating environmental compensation parameters; the moxibustion sense feedback input unit quantifies subjective moxibustion sense feedback of the patient; the syndrome differentiation and trend prediction unit is used for outputting dynamic syndrome type probability distribution and a syndrome evolution prediction sequence based on the traditional Chinese medicine knowledge graph; and the self-adaptive control decision unit is used for generating a final self-adaptive control instruction when the emergency safety condition is not triggered by integrating the environment compensation parameters, the moxibustion sense feedback, the syndrome type probability and the evolution prediction result. According to the system, three types of data sources are fused, a four-stage serial correction architecture is adopted, and multi-stage regulation and control are automatically completed in a treatment cycle, so that equipment is not out of alignment due to environmental fluctuation, downshift is performed no longer after scalding, and prescription change is performed no longer after symptom transformation, and therefore, temperature control precision improvement and manual intervention return-to-zero can be realized.

The invention discloses a rectal cancer postoperative recurrence risk prediction system and method based on multi-modal time sequence data, and relates to the technical field of medical artificial intelligence. The system comprises a data acquisition and preprocessing module, a feature extraction module and a multi-modal feature fusion and modeling module. The method comprises the following steps: constructing a cross-modal data set containing time sequence clinical data, a time sequence MR image and a biopsy digital pathological image; respectively extracting clinical features, radiomics and deep learning features of the MR image, and nucleus morphology and spatial distribution features of the pathological image; and fusing all the features by using a Transform network, and constructing a prediction model. According to the method, macroscopic images, micropathology and dynamic time sequence information are integrated, tumor heterogeneity is comprehensively quantified, the problem that prediction of a single-mode static model is not accurate is solved, the postoperative recurrence risk of the stage III rectal cancer patient can be evaluated more accurately, and clinical treatment decision making is assisted.

The invention relates to the technical field of medical time sequence data disease typing, and discloses a typing diagnosis system for diabetes mellitus. According to the system, time sequence data of a patient is generated through data preprocessing, segmentation is carried out through multi-scale sliding windows, statistical characteristics and trend slopes of physiological indexes are synchronously calculated in all the windows, and a mode track of healthy evolution of an individual is constructed. And dividing a diabetes subtype prototype by mapping the trajectory to a high-dimensional space. For a new data point, calculating a multi-dimensional distance and a migration probability from the new data point to each prototype, and generating a preliminary subtype membership degree vector; and feeding the vector back to a trajectory construction process, dynamically adjusting window parameters and calculating weights in an iteration mode, and outputting a stable personal health trajectory and an updated membership after optimization. And finally, the structured subtype diagnosis with confidence is generated through fusion. The system realizes dynamic, accurate and individualized typing of diabetes subtypes.

A new computational approach may provide improved detection of disease conditions and comorbidities, such as PTSD, Parkinson's, Alzheimer's, depression, etc. For example, in an embodiment, a computer-implemented method for detecting a disease condition may comprise receiving a plurality of data streams, each data stream representing a measurement of a brain activity comprising physical and chemical phenomena and performing pattern analysis on the plurality of data streams to detect at least one fundamental code unit of a brain code corresponding to a disease condition based on a combination of the plurality of data streams.

Disclosed herein is a method for providing dental recommendations based on image processing, in accordance with some embodiments. Accordingly, the method includes receiving a dental image of a patient from a device, analyzing the dental image using a machine learning model, identifying a dental anatomy of the patient based on the analyzing, identifying a dental pathology of the patient based on the analyzing, retrieving a dental reference dataset, processing the dental anatomy and the dental pathology with the dental reference dataset, generating a dental recommendation based on the processing, transmitting the dental recommendation to a dentist device and a patient device, and storing the dental anatomy dataset and the dental pathology dataset, receiving a campaign request from the dentist device, activating a targeted marketing for the patient based on the campaign request, identifying an advertisement for the targeted marketing based on the activating, transmitting the advertisement to the patient device.

The invention provides a breathing mode analysis method and device based on a calm breathing pressure difference waveform and electronic equipment, and relates to the technical field of medical signal processing and artificial intelligence, and the method comprises the following steps: obtaining a calm breathing pressure difference original signal of a target user; preprocessing the calm breathing pressure difference original signal to obtain a target waveform sequence; performing feature extraction based on the target waveform sequence to obtain a plurality of dimension features; combining the at least two dimension features to obtain a multi-modal feature; and inputting the multi-modal features into a trained classification model, and outputting a breathing pattern classification result. According to the method, the whole analysis process is completely based on natural calm breathing, special cooperation or forced breathing actions of the user are not needed, the pain point that traditional lung function examination has a high requirement for the user cooperation degree is solved, and auxiliary decision making of chronic respiratory diseases is achieved.

The invention discloses a digital management system for the health state of a patient after an anorectal operation, and relates to the technical field of digital management of the health state after the operation. Comprising a data acquisition and processing module which is used for calling an anorectal postoperative clinical path rule engine based on real-time symptom data input by a patient and the number of postoperative days, and dynamically generating and pushing a corresponding structured interactive data acquisition sequence; and the feature extraction module is used for carrying out real-time feature extraction on a postoperative recovery related image provided by a patient through a preset image analysis model at a patient terminal side so as to generate a multi-modal structured data set containing time sequence physiological indexes and image feature vectors. According to the method, clinical priori knowledge is fused through the multi-modal neural network, a compound and atypical early complication mode is deeply recognized from time sequence physiological indexes and image features, the early discovery rate of complication is increased, and the false alarm rate caused by conventional recovery variation is reduced through explainable risk contribution factor output.

The invention relates to a disease-specific quantitative trait locus identification method based on multi-omics integration. The method comprises the following steps: acquiring variation sites of whole genome sequencing data of a target object, and molecular phenotypes and molecular abundance of molecular phenotype data; determining an association significance probability value of an association pair formed by the variation point and the molecular phenotype based on the variation point and the molecular abundance, and screening a first association pair from the association pair based on the association significance probability value and condition analysis; determining a consistent second association pair in the normal association pair and the disease association pair, and determining a third association pair with a disease interaction effect in the second association pair; calculating a first effect estimation value and a second effect estimation value of each third association pair; and based on the first effect estimation value and the second effect estimation value of the third correlation pair, determining a target correlation pair related to the Parkinson's disease, and taking the target correlation pair as the identified quantitative trait site. By adopting the method, the Parkinson's specific pathogenic heritable variation can be accurately identified.

The invention provides a model for predicting lung adenocarcinoma prognosis and immunotherapy response based on copper death related LncRNAs, and belongs to the technical field of bioinformatics. According to the prediction model provided by the invention, the accuracy and the stability of prognosis judgment of a lung adenocarcinoma patient can be remarkably improved, and the defects of a traditional staging system in the aspects of reflecting tumor heterogeneity and individual treatment reaction difference are effectively overcome; the model not only can realize reliable individual risk grading, but also can further evaluate the tumor immune microenvironment state and predict the potential reaction of a patient to immunotherapy, so that a powerful auxiliary tool is provided for clinically formulating an accurate treatment strategy, in particular to the application decision of an immune checkpoint inhibitor; and finally, the method has important practical application value for improving treatment selection and life quality of patients.

The application discloses a kind of live cell dynamic monitoring method, comprising the following steps: obtaining the cell scanning picture group that live cell grown in culture vessel is photographed under specified condition, and the picture is preprocessed;Cell scanning picture group after pre-processing is input into trained cell identification model, and output cell growth state chart group;Cell growth state chart group is input into trained survival rate calculation model, and the survival rate of cell in each cell growth state chart is output;The cell survival rate numerical value obtained is drawn into cell survival rate-time curve chart;The cell survival rate-time curve chart generated is combined with nonlinear regression statistical method, and the half inhibitory concentration value of the efficacy of specified parameter calculation is selected.The growth state of cell under specified condition is monitored, and the survival rate of cell and the half inhibitory concentration value of the efficacy are calculated based on these data;It can significantly improve the efficiency and accuracy in the field of drug research and development, drug screening, treatment scheme effect, etc.

This invention discloses a method, device, electronic equipment, and medium for assessing the risk of pulmonary embolism. It relies entirely on collected non-invasive objective data and eliminates subjective judgments based on medical history or physician experience, effectively avoiding risk misjudgments caused by subjective variables in the PESI / sPESI model. This ensures the consistency and reliability of the assessment results. The ratio of Qa to Qb is used as the core assessment indicator. This ratio is a continuous variable, eliminating the need for empirical segmentation of parameters and avoiding significant differences in assessment results among patients with similar conditions. The risk level is assessed based on the ratio of Qa to Qb obtained through hemodynamic simulation. The results perfectly match clinical understanding of the severity of pulmonary embolism, improving assessment accuracy. Simultaneously, it reduces patient examination trauma and avoids high equipment and operational costs.

The application discloses a slice focal point map acquisition method and device, computer equipment and a storage medium, and relates to the technical field of artificial intelligence. The method comprises the following steps: acquiring a slice scanning image of a target sampling point of the slice through a digital slice scanner; performing defocus information prediction based on the image content of the slice scanning image to obtain defocus information of the slice scanning image; the defocus information is used for indicating the deviation degree between the axial position of the target sampling point and the focal plane position of the target sampling point; calculating the focal plane position of the target sampling point according to the defocus information; and generating the focal point map based on the focal plane position of the target sampling point. The application can enable the digital slice scanner to quickly generate the focal point map of the slice in the pre-scanning process of the slice, and the accuracy of the focal point map is relatively high.

This invention provides a method, device, and equipment for preoperative risk stratification assessment of endometrial cancer. It involves segmenting and annotating preoperative ultrasound images to extract radiomics feature vectors, and simultaneously performing structured encoding and mapping of pathological biopsy data to generate pathological feature vectors. These two feature vectors are then input into a fusion network containing a cross-modal attention gating module. This gating module automatically generates dynamic weight vectors based on the consistency between the image and pathological features. The two feature vectors are then weighted and modulated separately before being concatenated to obtain a joint feature representation. This automatically reduces the weight of the lower-confidence modality and amplifies the weight of the higher-confidence modality when their assessment conclusions are inconsistent. Finally, the joint feature representation is input into a multi-task prediction network to output the probability of myometrial invasion depth and the probability of lymph node metastasis risk, thereby generating preoperative risk stratification results and surgical plan recommendations.

The present disclosure relates to a drift detector to detect a drift in features of a diagnosis, the drift detector comprising a first collecting device configured to collect relevant explanations and raw data, a first database compiling the relevant explanations and raw data collected by the first collecting device, a second collecting device configured to collect relevant available expert knowledge from reliable sources, a second database compiling the relevant available expert knowledge from reliable sources collected by the second collecting device, a learned feature extractor configured to group initial data of the second database into a characteristic feature pattern for different diagnoses and to subsequently group the data from the first database into first feature patterns by diagnosis and from the second database into second feature patterns by diagnosis, a comparator configured to compare the empirical distributions of each of the grouped feature patterns in the databases with each other, and a processor configured to generate a report if results of the comparison do not comply with a predefined condition. Applications include medical applications such as recognizing new diseases.

To provide a new system and method for determining a primary site from a biomedical image.SOLUTION: As described herein. The computing system can identify a first biomedical image of a first specimen from one of a primary site or a secondary site associated with a condition of a first subject. The computing system can apply the first biomedical image to a site prediction model comprising a plurality of weights to determine a primary site of the condition. The computing system can store an association between the first biomedical image and the primary site determined using the site prediction model.SELECTED DRAWING: Figure 2

Described herein are systems and methods that related to techniques for managing individual data (such as patient data). In examples, systems can be configured to obtain data associated with individual profiles; obtain data associated with individual samples indexed in accordance with a period of time; and for each individual, link one or more entries of the individual profile corresponding to the individual with individual samples corresponding to the individual based on the period of time and the profile for each individual. In some examples, the system can generate workflow profiles corresponding to each individual of the plurality of individuals based on linking the one or more entries of the individual profile with individual samples. The system can then de-identify the workflow profiles corresponding to each individual of the plurality of individuals to generate limited workflow profiles that are used to test one or more models in a development environment.

This platform is a privacy-preserving, federated learning system designed to integrate whole-slide digital pathology images with matched molecular profiling and clinical metadata for improved diagnosis, subtyping, and prognostic estimation of gynecologic tumors. The architecture comprises local institutional nodes that retain raw patient data while participating in distributed model training coordinated by a central orchestration server. Each local node preprocesses whole-slide images into patch-level tensors, extracts visual embeddings via convolutional backbones, and processes molecular vectors (e.g., somatic mutations, expression summaries, copy-number measures) via a molecular encoder. A multimodal fusion module - implemented as an attention- based transformer - integrates image and molecular embeddings into a unified representation used by multi-task heads for classification (histologic subtype, diagnostic label) and regression (risk score). The federated learning controller aggregates encrypted model updates (FedAvg) and returns improved global weights without exchanging raw data, enabling cross-site generalization while preserving patient privacy. Explainability components generate attention maps and tile-level saliency (Grad-CAM style) linked to molecular features, providing interpretable morpho- molecular correlations to pathologists. The platform supports API integration with PACS / LIMS, conforms to privacy standards via optional differential privacy and secure aggregation layers, and is extensible to additional omics modalities or transfer / fine-tuning workflows for related tumor types. By combining multimodal fusion, federated training, and clinician-facing interpretability, the system accelerates robust, generalizable AI for precision pathology in gynecologic oncology.