52 results about "Stroke risk" patented technology

The present invention relates, in general, to perioperative bleeding and, in particular, to methods of identifying individuals at risk of perioperative bleeding. The present invention relates, in general, to perioperative renal dysfunction and, in particular, to methods of identifying individuals at risk of perioperative renal dysfunction. The present invention relates, in general, to perioperative stroke and, in particular, to methods of identifying individuals at risk of perioperative stroke.

Methods and apparatus are disclosed for the measurement of lipocalin-2 in body fluids (including but not limited to blood, serum, plasma, urine, saliva, tear, etc.) by an assay such as an immunoassay or an immunotest for (1) the prediction of risk of future cardiovascular diseases; and (2) the determination of the likelihood that certain individuals will benefit to a greater or lesser extent from the use of certain treatments designed to prevent and / or treat cardiovascular diseases.

The invention provides a thunder damage protection risk automatic assessment method for a power transmission line. The thunder damage protection risk automatic assessment method comprises the steps of collecting parameters of a typical pole tower and recording values of parameters comprising thunder parameters, pole tower parameters, landform parameters, line parameters and cloud-to-ground lightning parameters; establishing a back striking calculation model and a shielding failure calculation model according to the parameters of the typical pole tower; calculating the shielding failure trip-out rate n1 of the pole tower on the basis of electromagnetic transient calculation program and according to the parameters of the typical pole tower and the shielding failure calculation model; calculating the pole tower shielding failure trip-out rate n1=NgP1eta on the basis of a guiding developing model; calculating the average lightning stroke trip-out rate N according to the back striking trip-out rate n1 and the shielding failure trip-out rate n2; and performing grading on the whole line pole tower based on lightning stroke risks according to the average lightning stroke trip-out rate N. The pole tower with the severe anti-thunder situation can be found before lightning stroke faults occur in the line, modification measures are adopted ahead of time, and the lightning stroke trip-out rate of the power transmission line is reduced.

A system and method for medical image based patient-specific ischemic stroke risk prediction is disclosed. Left atrium (LA) and left atrium appendage (LAA) measurements are extracted from medical image data of a patient. Derived metrics for the LA and LAA of the patient are computed using a patient-specific computational model of cardiac function based on the LA and LAA measurements extracted from the medical image data of the patient. A stroke risk score for the patient is calculated based on the extracted LA and LAA measurements and the computed derived metrics for the LA and LAA of the patient using a trained machine learning based classifier, which inputs the extracted LA and LAA measurements and the computed derived metrics for the LA and LAA as features.

The invention provides a system and method for medical image based cardio-embolic stroke risk prediction. A system and method for medical image based patient-specific ischemic stroke risk prediction is disclosed. Left atrium (LA) and left atrium appendage (LAA) measurements are extracted from medical image data of a patient. Derived metrics for the LA and LAA of the patient are computed using a patient-specific computational model of cardiac function based on the LA and LAA measurements extracted from the medical image data of the patient. A stroke risk score for the patient is calculated based on the extracted LA and LAA measurements and the computed derived metrics for the LA and LAA of the patient using a trained machine learning based classifier, which inputs the extracted LA and LAA measurements and the computed derived metrics for the LA and LAA as features.

The invention discloses a distribution line flashover rate improvement algorithm based on an electrical geometric model, specific steps of solving the distribution line flashover rate by using the improved algorithm and considering factors such as lightning flashover density, tower shape, soil resistivity, line insulation level and other factors are proposed; the method comprises the following steps: determining line parameters, calculating a thunder-resistant level of direct lightning stroke, calculating equivalent flashover width of the line, and calculating the direct lightning stroke flashover rate and the inductive lightning flashover rate. According to the method, the influence of the lightning current amplitude on the equivalent lightning width of the line is considered based on theelectric geometric model, and the equivalent flashover width of the line is defined, so that the calculation of the distribution line flashover rate is more accurate, and the method has a certain guiding effect on lightning stroke risk assessment and differential lightning protection of the distribution line.

The invention relates to a method and a device of lightning stroke risk evaluation of a power transmission line. The method includes: acquiring a predetermined lightning damage membership degree function of each evaluation index of lightning stroke risks, and constructing a correlation degree matrix according to towers of a to-be-evaluated power transmission line area and the lightning damage membership degree function of each evaluation index; carrying out solving of correlation factors according to the correlation degree matrix, obtaining certainty degrees and uncertainty degrees of occurrence of lightning damage at the different towers according to the correlation factors, and obtaining basic probability distribution functions of the correlation factors; and using the basic probabilitydistribution functions as Mass functions in an evidence theory, and adopting the DS evidence theory to fuse the different correlation factors to obtain an evaluation result of the lightning stroke risks of the power transmission line. The method can increase an accuracy degree of lightning stroke risk evaluation of the power transmission line.

The invention discloses an early warning system for a stroke risk patient, and the system comprises a data obtaining unit which is used for obtaining the detection data, representing the stroke onsetrisk, of the patient; a data transmission unit which is used for uploading the detection data to the storage and calculation unit; a storage and calculation unit which is used for storing and preprocessing the detection data, and inputting the preprocessed detection data into a cerebral apoplexy disease deep learning model to obtain an output result which is cerebral apoplexy morbidity probability, thereby judging whether the output result exceeds a preset value or not: when the output result exceeds a preset value, an alarm signal is output; a display unit, which is used for displaying the detection data and a corresponding output result; an early warning unit, which is used for receiving and responding to the alarm signal. According to the invention, early warning can be carried out on the stroke risk in advance.

The invention provides a deep learning-based CBCT image cross-modal prediction CTA image stroke risk screening method and system. The method comprises the steps of 1, constructing a cyclic adversarialresistance generation network model; 2, training a cyclic adversarial resistance generation network model through the CBCT images and the contrast image data corresponding to the CBCT images; 3, inputting a test image into the trained cyclic antagonism generation network model to generate an angiography CT image; and 4, predicting the stroke risk according to the form, the carotid artery stenosisdegree and the curvature of the carotid artery in the angiography CT image. According to the method, based on the deep learning model, the non-enhanced CBCT image is converted into the enhanced CT angiography image, carotid artery blood vessel segmentation and extraction are carried out, the carotid artery stenosis degree and curvature are quantitatively calculated, then the stroke risk is predicted, and a convenient, economical and efficient new way is provided for clinically obtaining the CTA image and diagnosing.

The invention discloses an electric transmission line straight line angle tower lightning protection measure collocation method. The method comprises the steps that a line by-base tower is evaluated, and risk grades are divided; lightning protection modification m subgoals and weight coefficients of the subgoals are determined, and a goal weight column vector W is obtained; whether the tower lightning trip-out rate meets operating indexes or not is judged, and k alternative lightning protection measures are selected from common lightning protection measures; according to lightning protection modification actual investigation goals, the goal effect row vectors of the alternative lightning protection measures on the subgoals are determined one by one, k row vectors are combined to form a goal effect matrix F with dimensionality being k*m, and effect vectors of the alternative measures are obtained through calculation; the expected lightning trip-out rate after governance is calculated, and the lightning protection measure collocation process is conducted in an iteration mode. The electric transmission line straight line angle tower lightning protection measure collocation method is implemented based on lightning stroke risk evaluation results and combined with the lightning protection modification goals, optimizes lightning protection collocation, reduces the lightning trip-out rate, reduces the lightning protection engineering cost, lowers lightning protection construction difficulty and improves technology economy of the comprehensive lightning protection measures.

Three tier architecture for image-based diagnosis and monitoring application using Cloud is described. The presentation layer is run on the tablet (mobile device), while the business and persistence layer runs on a single cloud or distributed on different Clouds in a multi-tenancy and multi-user application. Such architecture is used for automated data mining application for computing (a) cardiovascular risk, stroke risk using IMT measurement, plaque characterization, (b) computing diagnostic index for benign vs. malignant tissue for ovarian cancer classification (c) benign vs. malignant tissue characterization for prostate cancer and (d) classification of fatty liver disease vs. normal cases. The Architecture is for data mining application.

The invention discloses a monitoring method of wind electricity unit thunder stroke times. The monitoring method comprises the steps that (1) temperature change values and temperature raising time data of all anti-thunder modules in a wind electricity unit are collected, and the state data values of the environment where the anti-thunder modules are located are collected; (2), whether the anti-thunder modules are struck by thunder or not is analyzed and judged in combination with temperature data signals of all the anti-thunder modules and state data signals of the environment where the anti-thunder modules are located collected in step (1), and the accumulated thunder stroke times of all the anti-thunder modules are calculated. The invention furthermore discloses a monitoring device of the wind electricity unit thunder stroke times. The monitoring device comprises the anti-thunder modules, a data collection module and a data processing module. The thunder stroke times of the anti-thunder modules are monitored through monitoring of the temperature change of the anti-thunder modules; and targeted anti-thunder design can also be achieved according to the collected data signals, impact and influences of a draught fan on a control system due to thunder stroke are reduced, the thunder stroke risk of the draught fan is lowered to a certain extent, and operation safety and stability of the wind electricity unit are improved.

It is the object of a method and apparatus for retinal vessel analysis based on digital images to enhance the ability to discriminate between healthy vessel states and at-risk vessel states while reducing manual effort and saving time in order to allow individual vascular risk, particularly stroke risk, to be determined in a more reliable manner and with fewer subjective systematic and random errors. The vessel segment diameter, type of vessel and the image coordinates are determined for a series of adjoining vessel segments along vessel portions in a measurement zone surrounding the papilla and are stored by vessel segment with reference to the evaluated image, to a reference image recorded with a time offset, and to a displacement vector that is determined for the vessel segment between the reference image and an evaluated comparison image. Comparison measurements are carried out only on identical vessel segments already measured in the reference image. The correlation of vessel segments to vessel portions and to vessel type is adopted intact from the reference image. The stored data sets for the vessel segments of the reference image and comparison images provide a progression of coordinate-oriented vessel segment diameters for all measured vessel segments as basis for determining parameters and presenting them in a spatially resolved progression, e.g., in progress images.

The invention provides a sleeping data analysis method for stroke prediction. The potential risk is found as soon as possible by analyzing the sleeping time, the sleeping efficiency, light sleep, deepsleep and other changes in the sleeping cycle and the sleeping stage of a participant, besides, after a prediction result is obtained with a common prediction model, the result is further processed in a similarity principle, so that the missing report rate is reduced, and the missing report rate is controlled in a receptible range. Compared with the prior art, continuous long-time sleeping data of the participator are acquired in a convenient and non-invasion manner, the stroke risk of the participator can be more accurately predicted in combination with part of known clinical physiological data. Meanwhile, by means of the method, the accurate rate of known prediction models can be increased by calculating the similarity of the participator with crowds suffering from stroke and crowds without stroke.

The invention relates to a rapid stroke identification method combining TOPSIS and artificial intelligence, and the method comprises the steps: firstly, judging whether the single-side muscle force decline exists or not through the horizontal lifting degree of the two arms of a user, and carrying out the classification of the muscle force difference through the angle between the arms; analyzing the degree of unclear mouth and teeth of the user through artificial intelligence speech recognition in combination with a word error rate calculation method; detecting position coordinates of face feature points by using an artificial intelligence face feature point recognition model, calculating a face droop degree through a TOPSIS method, and finally quickly recognizing the stroke risk degree through the TOPSIS method by synthesizing muscle force difference classification, an unclear degree of mouth and teeth and a face droop degree. According to the invention, the patient and the family members thereof can effectively and quickly identify the stroke symptom at home, and the stroke patient is automatically shunted to the corresponding stroke center for treatment according to different severity degrees of the stroke, so that the first-aid efficiency and quality of emergency medical aid service are improved.

The invention provides a microRNA marker combination for identifying the cause subtype of ischemic stroke. The microRNA marker combination comprises at least one of miR-125b, miR-125a, miR-let-7b, miR-let-7e, miR-7-2-3p and miR-1908. The invention further provides application of the microRNA marker combination in preparation of a reagent for identifying cause subtype of the ischemic stroke and application of the microRNA marker combination to preparation of a reagent for forecasting the stroke risk of a testee. The invention further provides a kit comprising a reagent capable of detecting thecontent of the marker combination. The microRNA marker combination provided by the invention for parting the ischemic stroke has the characteristics of simple and safe operation, high sensitivity, high specificity and easiness in massive parting.

Methods and apparatus are disclosed for the measurement of lipocalin-2 in body fluids (including but not limited to blood, serum, plasma, urine, saliva, tear, etc.) by an assay such as an immunoassay or an immunotest for (1) the prediction of risk of future cardiovascular diseases; and (2) the determination of the likelihood that certain individuals will benefit to a greater or lesser extent from the use of certain treatments designed to prevent and / or treat cardiovascular diseases.

The present invention relates to a method for predicting the risk of stroke of a subject and to a method for improving the prediction accuracy of a clinical stroke risk score. The methods of the present invention are based on the determination of the amount of Angiopoietin-2 (Ang-2) and / or the amount of Insulin-like growth factor-binding protein 7 (IGFBP7) in a sample from a subject. Moreover, thepresent invention pertains to the use of i) the biomarker Ang-2 and / or the biomarker IGFBP7, and / or ii) at least one detection agent that specifically binds to Ang-2 and / or at least one detection agent that specifically binds to IGFBP7 in a sample from a subject for predicting the risk of stroke of said subject.

The invention relates to the technical field of medical auxiliary systems, in particular to a stroke risk screening system and method. The method comprises the following steps: S1, a data processing step: acquiring and processing existing stroke medical data; S2, a model construction step: with the stroke medical data output in the S1 as a training set, performing iterative learning and training by using a risk prediction model constructed on the basis of an RBF neural network to obtain a prediction model; S3, a prediction step: collecting user data to be predicted, calling the prediction model, generating a prediction result and storing high-risk patient data to a medical data platform; and S4, a display step: displaying the prediction result in a visual form. According to the stroke riskscreening system and method provided by the invention, intelligent screening of patients with high risk of stroke can be realized, the accuracy and efficiency of diagnosis and treatment are improved,and intervention and examination recommendation in advance are facilitated. Meanwhile, screening and warehousing storage of high-risk groups can be achieved, digital management is facilitated, and early discovery, early warehousing, early management and the like are achieved.

The invention discloses a stroke risk assessment method and system based on related risk factors, and the method comprises the following steps: 1), collecting risk factor information and genetic information of an assessed person, and enabling the risk factor information to include personal information, disease history information, physiological and biochemical information and living habit information, wherein each type of information further comprising a plurality of information items, each information item can comprise one or more labels, and corresponding risk factor values (Value) are set for the labels and / or the information items, and (2) calculating a risk score f according to the risk factor information of the assessed person and the corresponding risk factor values of the genetic information.

The present invention relates to a fuzzy system that provides a measure of global cardiovascular risk based on the risk factors associated with cardiovascular disease and stroke. The fuzzy system is of multiple input single output (MISO) wherein the inputs are the values of risk factors and the output is the measure of global cardiovascular risk.

Disclosed herein are left atrial appendage (LAA) occluders that include self-powered physiological sensors to monitor physiological parameters of a subject. The sensors can be powered by harvesting energy generated by the patient's body or using wireless power delivery technologies. The disclosed devices can be used to close the LAA and to provide self-powering sensors to wirelessly monitor physiological parameters such as heart rate, pressure, temperature, size of the atrium, and levels of biomarkers such as C-reactive protein (CRP) and B-type natriuretic peptide (BNP) (e.g., using biosensors). In addition to addressing the stroke risk for patients with non-valvular atrial fibrillation, the disclosed devices offer post-surgical connected care that can reduce hospital readmissions, provide superior medical management, and improve patient quality of life.

The invention provides a method and system for stroke risk assessment in grass-roots public health services and a mobile platform. The method comprises the steps that 1 a health file corresponding tothe identity information of a resident is established; 2 the resident completes a physical examination and a questionnaire survey, and fills in a health data table; 3 all health data containing the items of interest in the health data table are entered; 4 a pre-set risk assessment algorithm is used to count the items of interest, and evaluate the risk level of stroke of the resident based on the statistical result; 5 initial screening conditions are set, and the resident who meets the initial screening conditions is marked as an object to be confirmed; and 6 the object to be confirmed is further diagnosed, and a doctor makes the final assessment conclusion of the risk of stroke. The method for stroke risk assessment in grass-roots public health services, which is provided by the invention,can overcome the technical problem that the prior art cannot make risk assessment when a stroke does not occur and intervene in high-risk groups in advance.

The invention discloses a system for prompting a stroke risk based on facial features, which obtains stroke treatment in advance by pre-judging the stroke risk. According to the technical scheme, thesystem comprises a memory, a processor and a terminal, the terminal comprises an image acquisition device, and the image acquisition device is used for acquiring face image information and transmitting the information to the memory; the memory stores the face image information transmitted by the image acquisition equipment and the data of a face feature model; and a detection analysis module is arranged in the processor, and is used for receiving the human face real-time image information and the data of the facial feature model sent by the memory, and judging whether the risk of stroke existsin the human face real-time image based on the facial feature model.

A composition and method for preventing or reducing the incidence of a transient ischemic attack in a subject at risk for developing a stroke comprises orally administering to the subject a prophylactically effective amount of a pharmaceutical composition comprising an ASIC la inhibitor capable of penetrating the blood-brain barrier. Preferred ASIC la inhibitors for use in the disclosed methods include amiloride and amiloride analogs.