102 results about "High risk group" patented technology

The invention discloses a sudden death risk real-time assessment system and method based on multi-modal physiological signal fusion, relates to the field of human physiological state monitoring and early warning, and solves the problems of low accuracy and poor real-time performance of sudden death risk assessment by single-modal signals. The system comprises a signal acquisition module, a preprocessing module, a high-dimensional feature extraction module, a multi-modal feature fusion module, a sudden death risk quantification module, a model updating engine module and an early warning feedback module, and each module integrates a self-adaptive filtering unit, a time sequence convolutional network unit, an improved multi-head self-attention mechanism unit and the like. According to the scheme, multi-mode signals such as electrocardio are synchronously collected through multiple channels, a dynamic risk index is calculated through self-adaptive noise reduction, parallel extraction of time-frequency domain nonlinear features and mutual information weighted fusion in combination with kernel density estimation, and online incremental updating and multi-stage early warning of a model are achieved; the sudden death risk can be accurately evaluated in real time, the anomaly detection sensitivity and the early warning timeliness are improved, and the method is suitable for daily health monitoring and high-risk group risk management and control.

The invention discloses a cerebral apoplexy high risk group positioning evaluation system based on multi-modal data, and relates to the technical field of medical information science, the cerebral apoplexy high risk group positioning evaluation system comprises a cerebral apoplexy prevention and control management platform, and the cerebral apoplexy prevention and control management platform is in communication connection with the following modules: a multi-modal data acquisition and integration module, the multi-modal data collection module is used for collecting multi-modal data related to cerebral apoplexy from multiple channels and carrying out preprocessing operation on the collected multi-modal data. By integrating clinical data, image data, omics data and terminal health data, multi-dimensional information related to the cerebral apoplexy can be comprehensively captured, particularly, cerebral vessel digital twin is utilized to simulate hemodynamic characteristics, a plurality of data sources are fused in combination with a graph neural network, high-risk groups can be recognized more accurately, and the accuracy of cerebral apoplexy recognition is improved. The accuracy and reliability of risk prediction are remarkably improved, the problem of missing detection caused by dependence on a single data source in a traditional method is solved, and more powerful support is provided for early intervention.

The invention relates to the technical field of medical informatization, in particular to an MCI cognitive trajectory and bidirectional outcome dynamic prediction method based on a joint model. The method comprises the following specific steps: firstly, constructing a univariate joint model, then constructing a multivariate joint model, and then constructing a function type multivariate joint model; performing parameter estimation and statistical inference on the model; fitting evaluation is carried out on the function type multivariable joint model based on multi-longitudinal cognition and multi-state outcome; and finally, carrying out individualized dynamic prediction. According to the method, a baseline or longitudinal nerve image is processed by using a function type data analysis technology, and a change track is reconstructed; a plurality of cognitive assessments are used as longitudinal measurement indexes, covariants such as age and gender and functional nerve images are used as predictive factors, two-way outcome of MCI and occurrence time of the two-way outcome are set as survival outcomes, a joint model based on multi-longitudinal cognitive multi-state outcomes is constructed, early screening of high-risk groups is assisted, and the survival outcomes of the high-risk groups are determined. And powerful support is provided for optimal distribution of medical and medical and health resources.

The invention relates to the technical field of medical data analysis and clinical risk prediction, provides a construction method of a glucocorticoid induced diabetes risk prediction model based on an LASSO algorithm, and belongs to the technical scheme of data processing executed on computing equipment. Clinical data of inpatients receiving systemic glucocorticoid treatment are collected, key variables are screened through LASSO regression, and a Logistic regression model is constructed to achieve risk prediction. The model is composed of four conventional clinical indexes, has good distinction degree and calibration degree, and finally realizes individualized risk assessment through column diagram form output. The method is simple, practical and accurate, and is suitable for SDM prediction and intervention management of clinical high-risk groups.

The invention provides a liver function data detection method and system for liver cancer high-risk groups, and relates to the field of health management. Context snapshots based on user IDs are created and updated in real time in a high-speed memory cache region, and a user diagnosis type, a medication state and a medication timestamp are combined to detect the liver function data of the liver cancer high-risk groups. According to the method and the device, the liver function data is subjected to differential judgment on the basis of the preset logic tree, so that the problems of misinformation and missing report caused by lack of context information in the prior art are effectively solved, the liver function data can be subjected to differential judgment accurately in real time according to the diagnosis type and the medication state of the user, misinformation and missing report are effectively avoided, and the user experience is improved. And the accuracy and the reliability of the alarm system are improved.

The invention discloses a breast cancer risk prediction method based on machine learning and multi-dimensional data, and belongs to the technical field of medical health information.The method comprises the steps that based on an NHANES database, diet, living habits and other information are collected, and a data set is formed; performing pretreatment; screening meaningful data features by using three feature selection methods of LASSO regression, mRMR and forward selection, and obtaining a final feature data set after intersection; dividing a training set and a test set; establishing a risk prediction model by using an SVM machine learning method, and learning the training set; performing model performance analysis on the test set to obtain a risk prediction probability of a final training model; the method has the advantages of multi-source data integration, high-precision prediction, personalized evaluation, dynamic updating and the like, risk factors of the breast cancer can be effectively mined, theoretical support is provided for prevention and treatment of the breast cancer, high-risk group screening is guided, morbidity reduction is assisted, early diagnosis and early treatment are achieved, and development of female health undertaking is promoted.

The invention discloses a heart and cerebral vessel early warning model based on vulnerable plaque image recognition. The model comprises the following steps: acquiring heart and cerebral vessel data of a patient; performing image preprocessing and data labeling on the cardiovascular and cerebrovascular data of the patient to obtain three-dimensional cardiovascular and cerebrovascular data, and establishing a vulnerable plaque recognition model by using a first convolutional neural network according to the three-dimensional cardiovascular and cerebrovascular data; extracting a three-dimensional blood vessel image, establishing a fluid simulation model for the three-dimensional blood vessel image, and acquiring blood flow information by using the fluid simulation model; according to the output of the vulnerable plaque recognition model and the blood flow information, obtaining rupture inducements and early warning information, inputting the rupture inducements and the early warning information into a second convolutional neural network model, and constructing an early warning model; detailed data information of the vulnerable plaque can be directly and conveniently obtained, and manpower and material resources are saved for diagnosis and treatment; the method provides accurate and powerful reference for doctors to diagnose and treat vulnerable plaques, and is beneficial to early discovery of high-risk groups with cardiovascular and cerebrovascular diseases.

The invention discloses a data mining technology-based construction method and application of an elderly diabetic hypoglycemia risk prediction model, and belongs to the technical field of medicines and computers. The method comprises the following steps: selecting elderly diabetic patients meeting standards, collecting demographic, disease and lifestyle factor data, dividing a training set and a verification set, comparing the performance of five data mining models, screening an optimal model, confirming the effectiveness of the model through external verification, and visualizing the result. The model can output a quantitative risk probability, is suitable for multi-scene real-time evaluation, can accurately identify high-risk groups, guides hierarchical intervention, reduces the incidence rate of hypoglycemia, and improves the management efficiency and safety of elderly diabetic patients.

The invention discloses a tumor radiotherapy rehabilitation nursing intervention system based on artificial intelligence, and relates to the technical field of tumor rehabilitation nursing. By integrating multi-dimensional data and combining clinical standards to output quantitative evaluation levels and key weak dimensions, rehabilitation weakness of a patient is accurately positioned, a targeted target is provided for the nursing intervention strategy generation module, a personalized scheme is generated, and pertinence and performability of intervention are ensured; policy execution data are captured in real time and a scheme is dynamically adjusted, so that the intervention process always adapts to the rehabilitation state change of a patient, and meanwhile, active management and control on high-risk groups are enhanced by periodically collecting core indexes and linking risk upgrading and priority intervention, so that the tumor radiotherapy rehabilitation nursing is changed from passive response to active pre-judgment, and the rehabilitation nursing efficiency is improved. And the overall quality of rehabilitation nursing after radiotherapy and the prognosis effect of a patient are optimized.

The invention discloses an enteral nutrition intolerance screening evaluation and management system, which belongs to the technical field of medical monitoring and comprises an information input module, a knowledge updating module, an information screening module, a risk scoring module, a scheme recommendation module, a feeding management module, a data analysis module, a user management module, a report generation module and a remote monitoring module. According to the method, the accuracy and reliability of risk assessment can be improved, the screening rate of enteral feeding intolerance high-risk groups is fully improved, the working efficiency of medical staff is improved, standardized and normalized feeding intolerance screening is achieved, the effectiveness and pertinence of treatment can be improved, it is ensured that recommended schemes are comprehensive and balanced, manual intervention is reduced, and the risk assessment efficiency is improved. Powerful decision support is provided for doctors, the actual requirements of patients are better met, the treatment effect and the satisfaction degree of the patients are improved, and the treatment compliance of the patients is enhanced.

The invention discloses an anterior cervical approach postoperation adjacent segment degeneration occurrence prediction model for a cervical spondylosis patient, and belongs to the field of cervical spondylosis postoperation prediction. The invention discloses a model for predicting occurrence of anterior cervical approach postoperative adjacent segment degeneration of a cervical spondylosis patient. The model comprises a classifier and risk factors input into the classifier, the classifier is used for preprocessing and sketching a region of interest by acquiring a medical image of a target region of a patient, then extracting image omics features, inputting the image omics features into various machine learning models to calculate a C index, and selecting an optimal machine learning model as the classifier based on the C index; the risk factors are obtained through screening by performing single-factor and multi-factor logistic regression analysis on clinical features. Through the model, a clinician can be helped to avoid related risks to the greatest extent when a cervical spondylosis operation scheme is made, meanwhile, high-risk groups with ASD after an operation are recognized, and a basis is provided for implementation of targeted follow-up visit and early intervention.

The invention discloses a differential privacy protection method and system based on dynamic gradient, and the method comprises the steps: 1, monitoring the gradient features of back propagation in real time in the training process of a target neural network model, and calculating a dynamic sensitivity zoom factor based on the gradient features; 2, identifying the type of the network hierarchy in the target model based on a three-level progressive judgment method; and step 3, setting a noise scaling factor according to the network hierarchy, selecting a noise type according to the intensity requirement of the network hierarchy for privacy protection, obtaining a noise scale corresponding to the noise type by using the dynamic sensitivity factor and the noise scaling factor, constructing noise according to the noise scale, and injecting the constructed noise into the corresponding network hierarchy to obtain the privacy protection. Obtaining a noise-added target model; and step 4, predicting input target data based on the neural network model after noise addition. According to the method, the protection effect of the model is improved, and stealing attacks aiming at genetic privacy and high-risk group identities can be effectively defended.

The invention discloses a system and equipment for predicting colorectal cancer and progression thereof. Based on the found expression level difference of AXIN2 and / or EPHB2 in colorectal cancer and the progression thereof, the invention provides a colorectal cancer and progression prediction method, system and equipment, a computer readable storage medium and a computer program product, high-risk groups can be efficiently and accurately screened out from common groups, and colonoscopy is further completed in time. While the economic burden caused by endoscopic screening is relieved, patients suffering from colorectal adenoma and early colorectal cancer are screened out, and the method has great significance in early diagnosis and treatment of colorectal tumor and further remarkable reduction of the morbidity and mortality of colorectal cancer.

This invention discloses an artificial intelligence-based intervention system for tumor radiotherapy rehabilitation nursing, belonging to the field of tumor rehabilitation nursing technology. By integrating multi-dimensional data and combining clinical standard output quantitative assessment levels and key weakness dimensions, this invention accurately identifies patients' rehabilitation shortcomings, providing targeted objectives for the nursing intervention strategy generation module and generating personalized plans to ensure the intervention's relevance and feasibility. It captures strategy execution data in real time and dynamically adjusts the plan, ensuring the intervention process always adapts to changes in the patient's rehabilitation status. Simultaneously, by regularly collecting core indicators and linking risk escalation and priority intervention, it strengthens proactive management of high-risk groups, realizing a shift from passive response to proactive prediction in tumor radiotherapy rehabilitation nursing, thus optimizing the overall quality of post-radiotherapy rehabilitation nursing and patient prognosis.

The invention discloses a biomarker combination for predicting PCI postoperative LVR of an STEMI patient and application of the biomarker combination, and belongs to the technical field of biology. The research discovers that in a left ventricular remodeling (LVR) group and a non-left ventricular remodeling group after percutaneous coronary intervention (PCI) of a patient suffering from acute ST segment elevation type myocardial infarction (STEMI), a left ventricular endocardial lower myocardial overall longitudinal strain (GLS internal) value, a lung B-line integral, a serum sFRP5 level and Apelin-13 show significant differences. Through combined prediction analysis, it is found that the GLS internal value, the lung B-line integral, the serum sFRP5 and the Apelin-13 have excellent prediction performance on whether the LVR occurs after the PCI operation of the STEMI patient or not, high-risk groups, suffering from the LVR after the PCI operation, of the STEMI patient can be accurately and efficiently recognized, intervention is conducted as soon as possible, and the life quality of the patient is expected to be improved.

The invention relates to the field of biological detection, and discloses a plasma polypeptide marker COL1A2 related to tumor bone metastasis and pain and application. Aiming at the problem that early diagnosis of bone cancer pain and tumor bone metastasis is difficult, by detecting the level of circulating COL1A2 in plasma, it is found that the circulating COL1A2 is remarkably up-regulated in patients with tumor bone metastasis and bone cancer pain and is remarkably related to clinical pain score and tumor staging. A sensitive and specific COL1A2 detection method is established by adopting an enzyme-linked immunosorbent assay (ELISA) technology, and occurrence and staging of bone cancer pain and tumor bone metastasis can be identified with high prediction efficiency that AUC is 0.95 and 0.94 respectively. The method is low in detection cost, simple and convenient to operate and suitable for clinical popularization and application, and scientific basis and technical support are provided for screening of high-risk groups, early diagnosis of bone metastasis, pain staging and personalized treatment. The invention has important significance for optimizing diagnosis and treatment strategies of patients suffering from bone cancer pain and improving clinical transformation efficiency.

The invention discloses an arteriosclerosis risk prediction method, device and system and a storage medium, and the method comprises the steps: obtaining a genetic risk score, a blood glucose index and basic features of a crowd arteriosclerosis index, thereby constructing an arteriosclerosis risk prediction model; and drawing a working characteristic ROC curve of the testee and calculating an area AUC under the ROC curve to evaluate the discrimination degree of the model for predicting the arteriosclerosis risk of the testee. By adopting the technical scheme of the invention, arteriosclerosis high-risk groups of the middle-aged and elderly people can be accurately screened out, and overhigh disease burden of arteriosclerosis of the middle-aged and elderly people can be prevented and controlled. Genetic and non-genetic factors are simultaneously included to judge the possibility of suffering from arteriosclerosis of middle-aged and elderly people, so that the method has certain practical guiding significance for preventing and controlling arteriosclerosis in people with abnormal blood glucose index levels.

The invention discloses a fracture risk prediction method and system, and relates to the technical field of health management. According to the prediction method, serious sarcopenia is selected to diagnose and assess the sarcopenia state, the ratio of monocytes to lymphocytes reflects the immune inflammation level, the prediction model is constructed in combination with the femoral neck bone mineral density T value, and the fracture risk influence factors are comprehensively covered; and the risk identification capability on special high-risk groups with sarcopenia, immune dysfunction and the like is effectively improved. And secondly, only three target independent variables are selected, and the three target independent variables are indexes which can be directly obtained or simply calculated in clinical routine examination, so that the data collection difficulty is greatly reduced. Meanwhile, the prediction model is visualized through a column diagram, complex calculation is not needed, a clinician can quickly inquire the fracture risk score through the actual value of the target subject, and the method is suitable for being efficiently applied to conventional outpatient service, hospitalization and other scenes.

The application discloses a CT dose self-adaptive optimization system and method thereof, and belongs to the technical field of medical images, which comprises a CT scanner, an image data processing workstation, a high-risk population analysis device, a dose parameter self-adaptive optimization device, a dose parameter automatic adjustment device and an intelligent learning unit, and based on the theory of information geometry, the dose parameter optimization problem is promoted from the Euclidean space to the Riemannian manifold space, the dose parameter manifold and the image quality manifold are constructed, the geodesic line is calculated on the manifold through the Fisher information matrix as the Riemannian metric, and the optimal dose path is found. The system adopts a double-manifold collaborative optimization mechanism, establishes a mapping relationship between the two manifolds, and realizes accurate optimization of parameters in combination with a geometric-guided genetic algorithm, and through pre-scanning, high-risk population analysis, information geometry optimization and automatic parameter adjustment, individualized dose control is realized, the radiation dose is additionally reduced by 30-45% compared with a traditional method, the image quality is maintained or improved, and the safety of CT examination is remarkably improved.

A novel DNA methylation marker TAGMe-3 capable of being used for tumor identification, and a use thereof. The TAGMe-3 gene sequence region has significant methylation differences between cancer tissue and paracancerous tissue, and it is only necessary for an abnormal high methylation state of the TAGMe-3 gene sequence region to be detected for a subject to be determined to be in a high-risk group for tumors. Moreover, the significant difference between tumor tissue and non-tumor tissue presented by TAGMe-3 is widely present in different types of tumors (pan-cancer).

This invention discloses a method for identifying intrahepatic cholangiocarcinoma (ICC) based on serum peptide characteristics and its application. The invention collects serum samples from three groups: individuals with ICC, those with benign liver disease, and healthy controls. Peptide characteristic peaks associated with ICC are screened, and a model is constructed using a logistic regression algorithm. This model demonstrates excellent discriminative ability on both the training and independent test sets, with AUCs reaching 0.986 and 0.963, respectively, significantly outperforming traditional tumor markers CA19-9 and CEA. The peptide characteristic peak combination and the discriminative model constructed based on it can be used for early detection of ICC, differentiation between benign and malignant cases, and screening of high-risk populations. It offers advantages such as rapid detection, standardized procedures, strong early ICC identification capability, and low false positive rate, providing a highly sensitive, specific, and widely applicable non-invasive screening tool for clinical use.

The invention provides a computing power device application technology for screening of liver cancer high-risk groups, early auxiliary diagnosis and postoperative treatment monitoring of patients, and belongs to the fields of biotechnology, medical diagnosis and treatment and massive health. By collecting samples of human body liquid such as peripheral blood, plasma, serum, saliva and the like and using biological and medical detection technologies, molecular markers (nationally approved that each kit is equal to the molecular marker) of a target gene methylation level, miRNA expression quantity and protein level or immune index quantity are detected; the problems of low sensitivity and poor accuracy in liver cancer high-risk group screening, early auxiliary diagnosis and patient postoperative treatment monitoring (including kits) are solved by using a computing power device application technology according to the detection result of any combination of the molecular markers, the precision can be greatly improved to prevent excessive medical treatment, and the purposes of precise medical treatment and clinical application are achieved. The method has important practical significance on screening of high-risk groups of liver cancer, early auxiliary diagnosis and postoperative treatment monitoring of patients, and has wide social benefits and use values.

The invention discloses a construction method and equipment of a medical service system for cerebral apoplexy prediction and analysis. The equipment comprises wearable hardware detection equipment, an intelligent terminal module, an intelligent AI module and a medical service module. Through the combination of the intelligent AI module and other modules, mass data can be integrated and analyzed, high-risk groups can be identified more accurately through deep learning of body health data of different individuals, different individual cerebral apoplexy prediction services are provided, and corresponding cerebral apoplexy prevention medical service schemes are formulated. Therefore, the medical service capability of early prediction and early prevention of the cerebral apoplexy is improved, the incidence rate and the recurrence rate of the cerebral apoplexy are more effectively reduced, the workload of medical care for cerebral apoplexy diseases can be greatly reduced through the intelligent AI module, and the accuracy rate of prediction of the cerebral apoplexy diseases is improved.

The invention belongs to the technical field of primary liver cancer diagnosis, and provides a primary liver cancer screening method based on quadruple biomarker detection, and the method comprises the following steps: collecting gender data, AFP data, AFP-L3 data, DCP data and CA199 data, and carrying out column diagram model matching; the construction process of the column diagram model comprises the steps of research patient determination, tumor biomarker data collection, liver function data collection, GALAD score and ASAP score calculation, statistical analysis, individual diagnosis biomarker screening, diagnosis value evaluation and column diagram construction. The invention provides a convenient and non-invasive detection means through the column diagram model constructed by the AFP data, the AFP-L3 data, the DCP data and the CA199 data, and improves the diagnostic performance of the primary liver cancer and the monitoring compliance of high-risk groups.

The invention provides target sections S6-1, S6-2 and S6-3 for identifying multiple cancer species and application of the target sections S6-1, S6-2 and S6-3. A group of unprecedented gene methylation sites are successfully identified through an innovative screening strategy and large-scale cross-cancer omics data analysis. The invention also comprises a detection reagent (such as a primer) and a kit for detecting the target segment of the genome. The methylation degrees of the target segment between a plurality of cancer tissues and para-carcinoma tissues have significant differences, so that the subject can be judged to belong to tumor high-risk groups. The cancer tissue and para-carcinoma tissue differential methylation state presented by the target segment can be widely applied to various different types of cancers.

The invention relates to a biomarker for diagnosing rheumatoid arthritis and application of the biomarker, and belongs to the technical field of biological medicines. The invention finds that the concentration of lipopolysaccharide and diamine oxidase in the plasma of a patient with rheumatoid arthritis (RA) is obviously higher than that of a healthy control group; furthermore, the invention also discovers that three proteins from plasma exosomes, namely phospholipid flip enzyme 1 (PLSCR1), keratin 1 (KRT1) and complement C1Q C chain (C1QC), have a good diagnosis effect on rheumatoid arthritis, and the diagnosis accuracy of rheumatoid arthritis can be remarkably improved by performing combined diagnosis on lipopolysaccharide and diamine oxidase with the three proteins. Therefore, when the biomarker is applied to preparation of RA diagnosis products, a core basis can be provided for accurate diagnosis of RA and generation of a risk assessment system, and the biomarker has transformation application value in the aspects of early diagnosis of RA and screening and accurate prevention and treatment of high-risk groups.

The invention provides a method for screening left atrial appendage thrombus by using a multi-mode ultrasonic technology, and relates to the technical field of medical ultrasonic diagnosis. According to the method for screening the left atrial appendage thrombus by using the multi-mode ultrasonic technology, through a layered screening process of noninvasive primary screening, precise secondary screening and comprehensive judgment, the technical advantages of four complementary modes of transthoracic ultrasound (TTE), transesophageal ultrasound (TEE), ultrasonic contrast (CEUS) and shear wave elastography (SWE) are integrated, primary risk stratification is completed by using noninvasive TTE, and secondary risk stratification is completed by using noninvasive TTE; tEE is combined with CEUS and SWE to achieve multi-dimensional evaluation of the structure, the blood flow and the hardness of the thrombus for high-risk groups, and finally accurate detection and stability evaluation of the thrombus are achieved based on the judgment standard of multi-modal data fusion. According to the method, the detection rate of the left auricle thrombus can be remarkably increased, the unnecessary invasive operation risk is reduced, meanwhile, thrombus stability basis is provided for formulating a clinical anticoagulation treatment scheme, and the method is suitable for conventional screening of thrombus high-risk groups such as atrial fibrillation and the like.

There is described herein a method of prognosing or classifying a subject with a Myeloproliferative Neoplasm (MPN) comprising: (a) determining the expression level of at least 10 genes in a test sample from the subject selected from the group consisting of SPP1, CEACAM6, GJA1, IGSF10, IGFBP2, COL4A5, LYVE1, MTIE, EMP1, XIST, DLK1, TPSAB1, TIMP3, CLC, MS4A1, ENKUR, ALOX12, KNDC1, HLA-DQB1, GAS2, CLEC2L, BEND2, CDH7, and NT5E; and (b) comparing expression of the at least 10 genes in the test sample with reference expression levels of the at least 10 genes from control samples from a reference cohort of patients; wherein a difference or similarity in the expression of the at least 10 genes in the test sample and the reference expression levels is used to prognose or classify the subject with MPN into a low risk group or a high risk group for worse survival.

The invention provides a breast cancer risk assessment method which is suitable for BI-RADS compact type high-risk groups with BMI greater than or equal to 28 and estrogen greater than 50pg / mL, and the method comprises the following steps: S100, injecting a temperature-sensitive nano contrast agent which is matched with fat acoustic impedance at 37 DEG C and has tumor environment impedance mutation at 39-42 DEG C; s200, dynamically adjusting the voltage of the molybdenum target tube according to the BMI and filtering by adopting a 40keV energy spectrum; s300, based on the leptin / adiponectin ratio L, the estrogen E and the focus area A, calculating a risk score according to R = 0.7 L + 0.3 E * (1 + 50A); according to the scheme, through high-risk group limitation, temperature-sensitive contrast enhanced ultrasonic specificity, BMI self-adaptive molybdenum target penetration and a multi-parameter fusion model, the detection precision is remarkably improved, the microfocus detection sensitivity reaches 92.3%, and the specificity reaches 88.7%.

The embodiment of the invention provides a high-risk group early warning method. The method comprises the following steps: acquiring a monitoring image in a target area; performing micro-expression recognition processing based on the monitoring image to obtain micro-expressions of all people in the target area; in all personnel in the target area, determining the personnel whose micro-expressions accord with high-risk tendency micro-expressions as target personnel; acquiring risk assessment information of the target person in a preset dimension, and performing high risk assessment on the target person based on the risk assessment information to obtain a risk assessment value of the target person; and based on the risk assessment value, determining whether the target person is a high-risk person in the target area. The relevance between the micro-expressions and the high-risk personnel is comprehensively considered, so that the accuracy is higher when early warning is performed on the high-risk personnel, and misjudgment can be avoided as much as possible when evaluation is performed in different dimensions.