55 results about "Predictive factor" patented technology

The invention discloses a power control method, system, equipment and medium for an optical storage and charging station, and belongs to the field of optical energy storage. A health degree factor is determined based on collected equipment operation data; determining a predictive factor based on the collected photovoltaic data; based on the power allocation weight and operation factors, a weight function is established to calculate the instruction weight of each device, and the operation factors comprise a planned maintenance factor, a sudden failure factor, an energy storage correlation factor, a health degree factor and a prediction factor of the parallel devices; calculating to obtain a power instruction of each parallel device; and controlling the parallel equipment to operate according to the power instruction through the station controller. According to the method, the bearable degree of the equipment is measured by calculating the dynamic weight comprehensively obtained on the basis of the health factor, the prediction factor, the maintenance factor, the sudden failure factor and the energy storage correlation factor, and the power instructions with inconsistent stagger degrees are distributed according to the actual operation condition, so that the operation stability is improved.

The invention relates to the technical field of biomedical engineering, and discloses a method and system for early warning and analyzing high-risk groups with high altitude polycythemia, and the method comprises the steps: screening a subject data set from an extremely high altitude area; performing structured processing on the subject data set to obtain a structured feature matrix, and extracting a core prediction factor from the structured feature matrix; candidate early warning models of the core predictive factors are generated, and the candidate early warning models comprise a logistic regression model, an XGBoost model and a random forest model; screening an optimal early-warning model from the candidate early-warning models, and establishing a high altitude polycythemia early-warning system of the subject data set through the optimal early-warning model; and effect verification is carried out on the high altitude polycythemia early warning system so as to realize high altitude polycythemia high-risk group early warning analysis processing of the subject data set. According to the method, the core problem that the early warning result is one-sided and unreliable due to three defects of data dimension missing, static evaluation limitation and extensive privacy mechanism can be solved.

The invention relates to a method and system for predicting post-induced hypotension of a patient, and belongs to the technical field of hypotension prediction.The method comprises the steps that original data reflecting the cardiovascular autonomic nerve regulation capacity of the patient is obtained; preprocessing the original data to obtain target data, and calculating a target pressure reflex sensitivity BRS and a target pressure reflex effective index BEI based on the target data and the same-direction change sequence of the continuous heartbeat; inputting the target pressure reflex sensitivity BRS, the target pressure reflex effective index BEI and the patient baseline data into a Logistic regression model to obtain a hypotension risk probability; and determining a hypotension risk level of the patient based on the hypotension risk probability and a preset threshold range. According to the method, the target pressure reflex sensitivity BRS and the target pressure reflex effective index BEI serve as predictive factors, so that the probability of predicting hypotension of a patient can be improved before the patient receives anesthesia induction.

The invention relates to a CVD risk assessment tool based on wearable device data and a machine learning algorithm, and the tool comprises a data collection module which obtains cardiovascular traditional risk factors through a cardiovascular risk assessment scale, and obtains monitoring data through a wearable device; the feature screening module is used for determining a candidate predictive factor range, and then performing feature extraction and predictive factor screening on the crowd wearing the wearable equipment by jointly using a minimum absolute contraction and selection operator (LASSO), a random forest (RF) and a Logistic regression model; the model building module is used for building a model based on an XGBoost machine learning algorithm; and the risk diagnosis module is used for outputting and evaluating the probability that the CVD risk of the individual is high within 10 years based on the characteristic data obtained by the wearable equipment data in real time. The cardiovascular health condition of an individual can be dynamically evaluated, personalized and timely risk early warning is provided, the method is suitable for daily health management, remote monitoring and other scenes, and a new solution is provided for early intervention and personalized medical treatment of cardiovascular diseases.

The invention relates to a method and a system for discriminating years of raw Pu'er tea in different storage aging periods. The method comprises the following steps: S1, preparing a sample; s2, analyzing components to determine the content of non-volatile compounds; s31, dividing the sample into three storage aging periods; s32, screening the number and the type of the non-volatile compounds by adopting LASSO regression to obtain a key predictive factor; s33, taking the key predictive factor as an input variable, and combining linear discriminant analysis to construct a year discriminant model; s4, outputting a result; the method has the advantages that based on LASSO regression of non-volatile compounds, linear discriminant analysis and other frontier machine learning technologies, an accurate model for Pu'er raw tea year discrimination is constructed, a matched visual module is combined, an efficient and visual tool is provided for Pu'er raw tea year discrimination, and the Pu'er raw tea year discrimination method is suitable for popularization and application. And scientific technical support is provided for quality evaluation and market transaction of annual Pu'er raw tea.

The invention provides a rubber graphene composite sealing element performance prediction method, which relates to the technical field of performance prediction, and realizes complete mapping from a bottom layer structure to a surface layer contact behavior by establishing a fine-grained structural unit, extracting node-level stress response data and coding the data into a contact response coding set; on the basis, joint matrix construction and regional response difference analysis are adopted, local response abnormal regions are effectively recognized, and the problem that interface microcosmic changes cannot be accurately captured through a traditional method is solved. Furthermore, a predictive factor sequence is formed through weighted combination of the response volatility and the contact strength, and a historical track is backtracked in a linkage manner, so that accurate early warning of a high-risk unit is realized.

The invention provides a construction method and a construction system of a chronic lymphocytic leukemia prediction model based on machine learning, electronic equipment and a storage medium, and relates to the field of chronic lymphocytic leukemia prognosis research. The construction method comprises the steps of obtaining original sample data, and performing preliminary screening; performing data cleaning on the screened sample data; determining a prediction factor from the plurality of features of the cleaned sample data; dividing the cleaned sample data corresponding to the prediction factor and the target variable into a training set and a test set; inputting the training set after unbalance processing into a LightGBM model for training to obtain a prediction model; inputting the test set into a prediction model, performing hyper-parameter optimization on the prediction model, and evaluating the performance of the prediction model; calibrating the prediction model to obtain a calibrated prediction model; the method has the beneficial effects that the method can be realized only by depending on conventionally available predictive factors, and the patient screening can be realized before the CLL clinical symptoms appear.

The present invention relates to a system and a method for monitoring mental health and generating personalized reports. According to one embodiment of the present invention, the method for monitoring mental health and generating personalized reports comprises: receiving a user's response data to a self-report questionnaire, wherein the self-report questionnaire comprises questionnaires corresponding to the respective mental health scales; and generating a report on the user's mental state based on the response data to the self-report questionnaire. The mental state report includes a personalized mental health graph that visualizes daily scores for each mental health scale in the form of a bar chart.The bar charts included in the personalized mental health graphic can be displayed in colors that are predetermined according to criteria for the respective mental health scale. According to the present invention, it is possible to capture psychological, biological, and social digital phenotypes of groups with non-suicidal self-harm via the self-report questionnaire and thereby investigate predictive factors related to these phenotypes.

The invention provides an extreme weather power load prediction method and system, and the method comprises the steps: obtaining historical data, and determining a basic load in extreme weather according to the historical data; constructing a total load decomposition model, stripping initial estimation of a basic load and a random load in the total load according to the total load decomposition model, and obtaining a meteorological load mid-value through multiple regression fitting; according to the meteorological load median, screening core meteorological factors through improved grey correlation analysis, and combining stepwise regression screening and extreme value adaptation correction to obtain an accurate meteorological load; constructing a comprehensive predictive factor set according to the precise meteorological load; and inputting each prediction factor of the comprehensive prediction factor set into the trained improved BP neural network model, and outputting a power load prediction value in extreme weather, thereby effectively improving meteorological load separation precision and extreme weather load prediction precision.

The invention relates to the technical field of data prediction, and discloses a colorectal progression stage adenoma data prediction method and system based on machine learning, a terminal and a storage medium, and the method comprises the steps: screening out a corresponding candidate prediction factor according to the colorectal progression stage adenoma data of each target sample; and dividing the candidate prediction factors into a training set and a verification set, constructing a target machine learning model by using the training set, inputting the verification set into a plurality of decision trees of the target machine learning model, outputting an original prediction probability, and converting the original prediction probability into a classification probability to obtain a prediction result. According to the method, an efficient, explainable and visual machine learning early warning model is utilized, a visual decision path is generated, conventional non-invasive indexes can be effectively integrated, and individualized prediction of colorectal progression stage adenoma data is achieved.

The invention discloses an ACS patient prognosis prediction method, which is characterized in that ACS patient prognosis is predicted through a trained and verified ACS patient prognosis prediction model, and prediction factors of the prediction model comprise patient age, heart rate, heart function grading, urea nitrogen and glycosylated hemoglobin. The predictive factors are screened by adopting multi-factor Cox regression analysis. The prediction model comprises a column chart model, and a column chart scale comprises a scale score range of 0-100, a patient age range of 25-95, a heart rate range of 40-140, a heart function grading range of I-IV, a urea nitrogen range of 0-35, a glycosylated hemoglobin range of 4-15, a total score range of 0-240 and a survival rate range of 0.99-0.01.

The invention relates to the technical field of silver powder production control, and discloses a production control method and system for silver powder and a storage medium. The method comprises the following steps: mounting a sensor array on a chemical reduction reaction kettle to collect data to form a process parameter set; filtering the parameter set to obtain a quality prediction factor; performing sectional temperature regulation and control based on the predictive factor, and dynamically controlling the reaction to obtain silver powder suspension; and inputting the suspension into a self-adaptive centrifugal system, and adjusting process parameters according to the sedimentation efficiency. The problem of multi-parameter collaborative optimization in a traditional silver powder production control method is solved, the problems that in the chemical reduction method preparation process, the concentration of silver ions fluctuates, the adding rate of a reducing agent is unstable, and the reaction temperature is difficult to accurately control are solved, and self-adaptive centrifugal separation and intelligent washing control in the silver powder post-treatment process are achieved.

This invention discloses a machine learning-based system for predicting mild cognitive impairment (MCI) and its progression to dementia. The machine learning-based MCI prediction system includes: selecting predictive factors from a population known to have mild cognitive impairment by screening candidate factors; using this population as a dataset and dividing it into a training set and a test set; inputting the training set into all machine learning models to train them, resulting in a pre-trained machine learning model; inputting the test set into the pre-trained machine learning model to test it, and using the machine learning model whose test index value exceeds a set threshold as the final trained MCI machine learning model; and inputting several predictive factors of the subject to be tested into the final trained MCI machine learning model to obtain a prediction result indicating whether the subject has mild cognitive impairment.

This invention relates to the field of intrinsic ability prediction technology and discloses a method for predicting the intrinsic ability characteristics of the elderly based on machine learning. First, intrinsic ability score data of the elderly in five dimensions, including cognition and psychology, are acquired, and heterogeneity classification is performed using latent profile analysis. Using the classification results as labels, random forest and LASSO regression are used in parallel to screen key predictive factors to form a feature subset. A classification prediction model is built based on XGBoost, and hyperparameters are optimized through cross-validation and grid search. The SHAP and LIME algorithms are integrated to achieve both global and local interpretability. Finally, the model is deployed on an online interactive platform, where inputting individual characteristics outputs the IC propensity classification, probability, and feature contribution explanation. This invention achieves accurate classification and interpretable prediction of the intrinsic abilities of the elderly, reduces the dependence of assessment on professionals and the environment, and is suitable for large-scale application in grassroots elderly care scenarios.

This invention discloses a machine learning model integrating oral microbiome and clinical features and its application in predicting sarcopenia at high altitudes. The method targets high-altitude populations residing at altitudes >3500 meters, collecting oral microbiome 16S rRNA sequencing data, clinical features, and lifestyle data. Core predictive factors are selected through adaptive preprocessing and elastic network regularized regression, and a classification model is constructed using logistic regression. To address the sample imbalance problem in high-altitude areas, this scheme introduces a synthetic balanced sampling strategy based on target distribution doubling; simultaneously, Platt Scaling is used for two-layer probability calibration, combined with a dynamic threshold optimization mechanism based on F1-Score to improve discrimination accuracy. This invention is the first to integrate oral microbiome shaped by the high-altitude environment (such as…)… Selenomonas , Veillonella (etc.) are incorporated into the predictive model. Experiments show that the model achieves an AUC of 0.868 on the independent test set and has advantages such as being completely non-invasive, low-cost, and highly interpretable. This invention provides an efficient early screening program for sarcopenia in resource-scarce high-altitude areas.

The invention provides a dilated cardiomyopathy child death risk prediction method, system and device, and relates to the technical field of medical prediction models.The method comprises the steps that clinical data of a target dilated cardiomyopathy child patient are obtained and input into a pre-constructed nomogram prediction model, and death risk prediction results of the target dilated cardiomyopathy child patient at multiple time points in the future are output; the nomogram prediction model is constructed through independent prediction factors, the amino-terminal brain natriuretic peptide precursor level, the gender and the digoxin use condition which are determined through single-factor and multi-factor Cox regression analysis. According to the method, the nomogram prediction model is constructed by integrating the key clinical variables, accurate prediction and individualized treatment guidance of the dilated cardiomyopathy death risk of children are achieved, and the accuracy of clinical decision and the prognosis management effect are effectively improved.

The invention relates to the technical field of medical artificial intelligence, and particularly discloses a method for establishing a disease prediction model of GCK-MODY in gestational diabetes population based on machine learning. The model takes five clinical characteristics as input variables: a body mass index before pregnancy, fasting blood glucose and glycosylated hemoglobin levels during GDM diagnosis, continuous multi-generation diabetes family history and a newly discovered predictive factor, namely GDM diagnosis week of pregnancy. A classification model is constructed through a support vector machine algorithm, and a SHapley additive interpretation method is adopted to provide global and local interpretation of a prediction result, so that model transparency is enhanced. The area of the model under a curve in a test is obviously superior to that of an existing screening standard. The invention also provides an electronic device, a storage medium and a prediction system comprising the model, which can assist clinicians in early recognition of GCK-MODY high-risk patients, provide decision support for targeted gene detection, avoid unnecessary enhanced hypoglycemic treatment of pregnant women, reduce the risk of maternal and infant complications, and promote precise medical development of gestational diabetes mellitus.

The invention discloses a pregnancy individualized medication dosage prediction factor screening method, a prediction model construction method and a prediction system, and relates to the technical field of medicine. Factors which are obviously related to blood concentration and have no obvious multicollinearity among variables are screened out from factors influencing pregnancy drug dosage to serve as drug dosage prediction factors, and a corresponding prediction model and system are constructed. According to the screening method, the prediction model construction method and the prediction system provided by the invention, auxiliary decision making can be provided for individualized drug dosage during pregnancy.

The invention belongs to the technical field of medical risk assessment, and provides a fibrosis metabolism related steatohepatitis risk assessment system. Comprising a data acquisition module configured to determine candidate prediction indexes; the primary screening module is configured to perform single-factor logistic regression analysis on the candidate prediction indexes to determine key prediction indexes of fibrosis metabolism related steatohepatitis; the secondary screening module is configured to screen out four independent predictive factors from key predictive indexes of fibrosis metabolism related steatohepatitis through stepwise multiple regression analysis; and the model training module is configured to train the constructed multi-factor logistic regression model based on the screened four independent predictive factors. According to the method, four indexes of aspartate transaminase, alanine transaminase, triglyceride and high-density lipoprotein cholesterol are integrated, a fibrosis MASH evaluation and prediction model for obese people is constructed, and the evaluation performance is better.

This invention discloses a method and related equipment for predicting and intervening in the risk of stillbirth related to umbilical cord torsion. The method acquires multimodal clinical data, including the pregnant woman's complaints and prenatal ultrasound examination information, and standardizes the data to extract core predictive factors such as changes in fetal movement, signs of umbilical cord root torsion, and fetal growth restriction, as well as auxiliary predictive factors such as abnormal amniotic fluid volume and abnormal placental-umbilical cord insertion, constructing a feature vector that the model can process. Based on a pre-trained risk prediction model, the risk of stillbirth related to umbilical cord torsion is quantitatively assessed, a comprehensive risk score is output, and risk levels are classified. Based on the risk level, a pre-set clinical intervention strategy is automatically matched to form standardized treatment recommendations, which are then output to the clinical terminal. This method overcomes the limitations of traditional methods that rely on low sensitivity of single ultrasound diagnosis, achieving early risk identification and tiered management.

This invention discloses a method for predicting the glycemic index (GI) of oat products in vitro, belonging to the field of food testing technology. This invention systematically integrates multi-dimensional data on the nutritional components, structural characteristics, and in vitro digestion kinetics of oat products. Using machine learning algorithms such as Bayesian ridge regression, it screens out five key predictive factors: β-glucan content, damaged starch content, median particle size, aleurone layer thickness, and 180-minute hydrolysis index. Combined with in vivo GI measurements, a high-precision prediction model is constructed through nonlinear regression fitting. This invention transforms the complex GI formation mechanism into a clear and measurable quantitative relationship, providing an efficient and low-cost technical solution for the rapid development, process optimization, and quality control of low-GI oat products, and has clear industrial application value.

The invention provides a 30-day death rate prediction device and method for cerebral hemorrhage patients, equipment, a medium and a program product, relates to the technical field of medical biology, and analyzes data in MIMIC and eICU-CRD databases. And performing feature selection by using Lasso regularization, mutual information and recursive feature elimination (RFE). And training to obtain a prediction model, and performing internal verification and external verification by applying comprehensive performance indexes. The feature importance is interpreted using an SHAP value. Determining key predictors includes acute physiological score III (APSIII), age, leukocyte (WBC), and albumin level. Compared with traditional scoring systems such as GCS and APSIII, the model shows higher prediction accuracy in the aspect of predicting the death rate in 30 days, and has more balanced sensitivity and specificity. The machine learning method provided by the embodiment of the invention shows a promising prognosis value in the aspect of predicting the 30-day death risk of ICH patients received by ICU. It needs to be noted that the models can only serve as supplementary judgments and cannot replace clinical judgments.

The present invention relates to a method for providing information for predicting therapeutic responsiveness or prognosis to transarterial chemoembolization. The gut microbiota of the present invention serves as a biomarker for predicting therapeutic responsiveness or prognosis to transarterial chemoembolization in patients with liver cancer. The gut microbiota of the present invention can be utilized as a potential predictive factor for clinical outcomes of transarterial chemoembolization, can also be used as an adjunct therapeutic target for treatment of liver cancer patients, and enables prediction of survival rates of liver cancer patients who have undergone transarterial chemoembolization. Accordingly, the biomarker of the present invention is expected to be usefully applied as a marker for predicting therapeutic responsiveness or prognosis to transarterial chemoembolization in patients with liver cancer.

The invention relates to the technical field of power grid loads, in particular to a power grid load prediction method and system, and the method comprises the following steps: analyzing the dominant influence of parameters such as air temperature, wind speed and rainfall on a load based on the node load and regional meteorological data of a power grid, calculating a meteorological and load cooperation relation in combination with a user time-sharing power consumption behavior label, and correcting a prediction trend. And obtaining a load prediction updating result. According to the method, node meteorological parameters and load changes are synchronously mapped through multi-period association analysis and a hierarchical feature screening strategy, deep description of load fluctuation driving factors is realized, a label classification mechanism of user behaviors enables power consumption modes in each period to be accurately expressed, richer behavior dimension support is provided for parameter prediction, and the prediction accuracy of the power consumption modes is improved. Self-adaptive evolution of predictive factors is achieved through periodic feedback adjustment, the weight is corrected in real time, and the recognition and prediction capacity on extreme scenes such as sudden change and fluctuation is remarkably improved.

This invention relates to the field of medical prediction technology, specifically to a method and system for constructing a prediction model for the occurrence of TMA (tumor homicide) in pediatric HSCT patients. The method involves acquiring sample data from each pediatric patient and grouping them; encoding variables according to coding rules; comparing the differences in variables between the TMA group and the non-TMA group to obtain the inter-group difference test results for each variable, thereby screening candidate variables; using the LASSO algorithm with an L1 regression model penalty term to screen the candidate variables and obtain key predictive factors; combining a logistic regression model and multiple machine learning algorithms into an initial occurrence prediction model and training it to obtain the final occurrence prediction model. The model constructed by this technical solution can predict based on the variables corresponding to the key predictive factors, and since the model is trained using pediatric sample data, it fully considers the special characteristics of pediatric patients, resulting in high accuracy in risk assessment.

The invention provides application of 73rd histidine methylation of serum actin as a marker in preparation of a reagent for assisting early prognosis evaluation of sepsis, and belongs to the technical field of biomarkers. Tests show that the 73rd site histidine of serum actin is subjected to methylation modification, and the differences among a Control group, a Survivor group and a Non Survivor group are remarkable; the secondary spectrum of the target peptide fragment shows that the peptide spectrum matching effect of the modified peptide fragment is good, and the modification identification is credible; and the area under the ROC curve and the 95% confidence interval are 73.56% (60.17%-82.68%). Under the optimal cutoff value of 30.29, the sensitivity is 0.792, the specificity is 0.639, the positive predicted value is 0.656, the negative predicted value is 0.780, the positive likelihood ratio is 2.198, and the negative likelihood ratio is 0.325. Therefore, the 73 histidine methylation of serum actin is an independent predictive factor of sepsis prognosis, can be used as a reliable and potential biomarker for early prognosis evaluation of sepsis, and has important clinical application value.

The invention discloses application of uridine as a serum marker in prognosis of hepatitis B related chronic-acute hepatic failure, the uridine content is an independent predictive factor of 28-day mortality of patients with hepatitis B related chronic-acute hepatic failure, and when the mortality of the patients is increased, the uridine expression quantity is reduced, so that the content of the uridine in the prognosis of hepatitis B related chronic-acute hepatic failure is reduced. Uridine can be used for constructing a hepatitis B related chronic-acute liver failure prognosis prediction model. The serum marker uridine provided by the invention can be used for identifying high-death-risk patients in the early stage of hepatitis B-related chronic-acute hepatic failure, provides important reference for clinical doctors to accurately judge the severity of illness, and can identify the high-death-risk patients in the early stage and take corresponding positive treatment measures.

The invention relates to the technical field of elevator dispatching optimization, in particular to an elevator group control energy-saving dispatching method and system based on load prediction.The method comprises the steps that the actual passenger flow with the same time property as the current moment in elevator historical operation data is analyzed, and a first predictive factor is constructed; the method comprises the following steps: acquiring a pedestrian flow load, taking the pedestrian flow load as a reference prediction value, analyzing the similarity between the reference prediction value and all historical operation data, distributing contribution degree weight to the historical operation data of each day, constructing a second prediction factor, obtaining a pedestrian flow prediction value, and finally correcting the pedestrian flow prediction value in real time according to the difference between the pedestrian flow prediction value and the actual pedestrian flow load of the current day; and constructing an adaptive inertia weight of a particle swarm algorithm based on the human traffic prediction correction value, and searching an optimal solution to perform response scheduling. The invention aims to improve the dispatching efficiency of the whole elevator group.

The invention discloses an ankylosing spondylitis spinal injury severity prediction model construction method and system, and the method comprises the following steps: S1, carrying out the data collection of an ankylosing spondylitis patient, the data including epidemiological data, clinical phenotypes and laboratory detection indexes; s2, layering the risk layers, and dividing a spinal injury severity high-risk group and a spinal injury severity low-risk group by taking mSASSS = 25 as a threshold value; s3, eliminating radiology dependent variables through logistic regression analysis, and screening out independent predictive factors; s4, constructing a spinal injury severity prediction column graph model based on the independent prediction factors; s5, vertically mapping the measured value of each index of the patient to a scale axis corresponding to the column diagram to obtain a single score; and after the total score of the six items is accumulated, positioning is performed on a risk coordinate axis to obtain a predicted risk probability. The method improves the disease prediction capability, is high in clinical universality, is convenient and efficient to operate, and has a treatment intervention guidance value.

The invention relates to a mild cognitive impairment reversal prediction method and system based on a multi-modal user portrait, and the method comprises the steps: obtaining multiple types of data of a mild cognitive impairment patient through multi-source data collection, and constructing the multi-modal user portrait; carrying out standardized preprocessing on the data, and carrying out probabilistic clustering by adopting a Gaussian mixture model to divide patient subtypes; screening predictive factors through single factor analysis and LASSO regression, and performing oversampling on the training set by adopting an SMOTE-NC algorithm to balance data distribution; a prediction model is constructed based on XGBoost, and model hyper-parameters are optimized through grid search and cross validation; after the construction and training of the prediction model are completed, the prediction model is used for carrying out reverse prediction on newly-collected multi-modal user portrait data of the mild cognitive impairment patient; meanwhile, the contribution degree of each feature in the prediction model to a prediction result is analyzed based on the SHAP value, so that the interpretability of the model is enhanced. According to the method and the system, prediction errors can be reduced, and the accuracy of predicting mild cognitive impairment reversal is improved.