58 results about "Single factor analysis" patented technology

The invention discloses a multi-scale information integration tight formation brittleness index measuring method. The connotation of a brittleness index is clear based on rock mechanic and acoustic analysis, and the brittleness index is redefined and a brittleness index calculation model based on acoustic logging data is established; on the basis, rock composition and mineral content are obtained in combination with X ray diffraction experiments; the brittle mineral type of a researched target region is obtained based on single factor analysis and a new brittleness index calculation formula is established. Through the analysis, a formation composition profile is obtained by processing routine logging data so as to achieve refined modeling and continuous processing of the brittleness index. According to the measuring method disclosed by the invention, the brittleness index measurement can be achieved in wells in shortage of special logging data such as array acoustic wave so as to avoid measurement errors caused by transverse wave prediction and the like, exploration cost can be saved and the explanation precision can be improved; therefore, the measuring method has significance on fracturing transformation and sweet point searching in a tight formation.

The invention discloses an establishment method of a solitary pulmonary nodule malignancy probability prediction model. The establishment method particularly includes the steps: acquiring basic information of patients and serum tumor marker levels 1-7 days before operation; dividing patient cases into one group with GGO (ground glass opacity) lesion proportion higher than or equal to 50% and another group with GGO lesion proportion lower than 50% according to the GGO lesion proportion and CT (computed tomography) imaging reports of the patients; setting experiment groups and validation groups in each group of cases according to the proportion of 3:1, performing single-factor analysis on relative data of cases of the experiment groups to initially screen independent risk factors; substituting the independent risk factors into multifactor analysis to obtain independent risk factors for judging benign and malignant SPNs (solitary pulmonary nodules); acquiring the SPN malignancy probability prediction model by the aid of Logistic regression; substituting case data of the validation groups into the model, and verifying the case data of the validation groups. The model is simple and easy to use, used indexes can be acquired by the aid of routine examination and are easy to use, and effective intermediate reference information can be provided for further diagnosis and treatment of doctors according to the model.

The invention discloses a value analysis method of underpinning pile foundation influenced by up-down overlapping tunnel shield tunneling; the method comprises the following steps: using FLAC 3D finite difference software to simulate an underpinning pile foundation model bearing load, and parsing settlement and deformation rules; aiming at an overlapping tunnel excavation process after pile foundation underpinning, and simulating overlapping tunnel excavation by shield tunneling method; parsing ground surface settlement and pile foundation displacement stress influenced by the shield tunneling method; carrying out single factor analysis contrast for physical parameters including excavation sequence, tunnel spacing, and underpinning pile foundation deep burying, pile diameter and spacing; analyzing different factor influences on the underpinning pile foundation, thus finally obtaining the corresponding simulation value and pile foundation settlement and deformation trends. The value analysis method is creative, high in accuracy, and thus providing wide engineering application prospects.

The invention discloses a construction method of a decompensated liver cirrhosis combined infection risk prediction model. The method comprises the following steps: S1, data acquisition: collecting decompensated liver cirrhosis patient information; S2, data preprocessing: cleaning and sorting the original data; S3, index screening by adopting LASSO regression: dividing the patients into an infection group and a non-infection group according to whether the patients have combined infection or not, and performing single-factor analysis on the grouped index data of the patients to obtain single-factor meaningful indexes; bringing the single-factor meaningful indexes into Lasso regression for index re-screening to obtain indexes for constructing a prediction model; and S4, prediction model construction: constructing the prediction model by using the indexes screened by Lasso regression through multi-factor Logistic regression. The method is based on application of a clinical big data methodand is high in reliability; the constructed model is simple and easy to use, and the used indexes can be obtained through conventional inspection and are easy to obtain.

The invention provides a method for evaluating the usability of eye movement technology based medical device manual, which utilizes a Tobii eye tracker to obtain the eye movement data of a subject who reads two kinds of medical device manuals. Five kinds of eye movement data associated with the usability evaluation are selected for analyzing cognitive activities and as usability indexes. According to the relationship between the five kinds of eye movement data and cognitive activities and based on three criteria of attraction, search speed and search accuracy, the using effect of the manual is evaluated. The questionnaire is used to evaluate the rationality of an eye movement model. The variance of five kinds of eye movement data under different areas of interest are analyzed by single factor analysis of variance. A multiple comparison method is used and comprehensive analysis is done in combination of the evaluation results of the eye movement model so as to find the influence of different types of interest areas on the eye movement data and put forward suggestions on usability improvement. The method, with high accuracy and quantitative characteristics, has strong reference value for the design and improvement of such manuals.

The invention discloses a coal-bed gas well multi-coal seam combined drainage and mining technique evaluation method and belongs to the technical field of coal-bed gas exploitation. According to coal-bed gas geological exploration data and historical data, the method carries out comparison and analysis on reservoir parameters among a plurality of coal seams in a multi-coal seam region and combined-layer drainage and mining adaptation evaluation. Under an evaluation system above, the method comprises reservoir parameter single factor analysis, parameter weight assignment and final multi-parameter comprehensive assessment, and finally, realizes the purpose of judging whether target coal seams are suitable for combined-layer drainage and mining.

The invention relates to a biomarker used for prognosis of lung cancer. 95 lung cancer prognosis specific genes are selected from 2738 lung cancer gene expression programming data from various countries through statistical analysis, wherein 85 of the selected genes are related to survival probability and postoperative relapse probability that reported for the first time. Furthermore, the 95 lung cancer prognosis specific genes are subjected to single-factor analysis, multi-factor analysis and survival curve analysis to finally accurately select seven representive genes. Clinical data in various countries proves that the gene markers can be very effectively used for predicting the survival probability and postoperative relapse probability of lung cancer, and according to this, lung cancer patiens are divided into low-, medium- and high-risk groups.

The invention discloses an intelligent geological disaster risk analysis method, comprising S100, establishing an ArcGIS independent database of the cloud under the monitoring network, and establishing a unified data interaction language, S200, collecting and integrates regional geological data through a data integration program, and forming a data dictionary description of the data, uploading thedata dictionary description of the data to the ArcGIS independent database through a data integration system, and backing up the data to an emergency database, S300, performing single factor analysis, multiple factor analysis and multi-information mixed factor analysis moduling through ArcGIS, and forming an information volume model, and S400, building a data analysis model program in a GIS database, forming a multi-level data analysis layer, and displaying data on an ArcGIS platform. The intelligent geological disaster risk analysis method provides more effective and intelligent data integration and data analysis, and therefore a more specific risk analysis result is obtained.

The embodiment of the invention provides a model training method and related equipment, which are used for improving the accuracy of a model and reducing the model training time. The method comprisesthe steps that N prediction topics are determined, and N is a positive integer larger than or equal to 1; factor labeling is carried out on the data corresponding to the N prediction themes to obtainN groups of labeling factors, and the N prediction themes correspond to the N groups of labeling factors; the N groups of labeling factors are preprocessed to obtain the N key data sets; the single factor analysis is performed on each key data set in the N key data sets to obtain N groups of target factors, the N groups of target factors having an association relationship with the N prediction themes; and training the N groups of target factors through at least one model algorithm to obtain at least one model corresponding to the N prediction topics.

The invention discloses a hierarchical important feature selection method based on clinical high-dimensional breast cancer data. The feature selection method comprises steps of statistical feature selection and integrated feature selection. At the statistical feature selection step, a single factor analysis method is employed; and features having obvious impacts on outcome variables are selected preliminarily by different statistics and checking. At the integrated feature selection step, a gradient lifting tree is established; after model training, a feature importance score is obtained; witha designed and verified importance score threshold, selection of features having obvious impacts on outcome variables is realized. Therefore, problems of too high data feature dimension, excessive redundant features and data disorder during the clinical breast cancer prediction modeling process are solved effectively; the redundant or meaningless features in high-dimensional clinical breast cancerdata can be excluded; and thus a few of features having the important impact on the breast cancer modeling are selected and the high accuracy and practicability of the breast cancer model are ensured.

The invention belongs to the technical field of medical data analysis and relates to a method for analyzing rating parameters of relevant factors of diseases, in particular to a method for analyzing overall rating parameters of sarcoidosis and atypical tuberculosis. The method for analyzing the overall rating parameters of the sarcoidosis and the atypical tuberculosis disclosed by the invention comprises the following steps of: selecting patients with the sarcoidosis and the tuberculosis, who are confirmed by a pathologic biopsy and a treatment follow-up visit, by using a clinical epidemiology case-control study method, so as to carry out retrospective single-factor analysis; selecting variables of danger factors, clinical manifestations and characteristics of iconography and pathology, which can identify the sarcoidosis and the tuberculosis, to carry out Logistic regression analysis by combining with the conventional technical standard; calculating weight fractions of all the variables and establishing a rating model; and analyzing the overall rating parameters of the sarcoidosis and the atypical tuberculosis by using model software. The method disclosed by the invention can supply important references and a rapid and efficient auxiliary analysis for clinical application.

The invention discloses a method for constructing a lymph node metastasis prediction model of a breast cancer patient based on radiomics. The method comprises the following steps: acquiring magnetic resonance image data and clinical feature data of the patient; extracting image features based on the magnetic resonance image data; screening the image features by using a random forest algorithm to obtain a plurality of key image features, and establishing an image feature prediction model based on the key image features by using a support vector machine algorithm; performing single-factor analysis screening on the clinical feature data to obtain key clinical features, and establishing a clinical feature prediction model according to the key clinical features by adopting a support vector machine algorithm; and establishing a lymph node metastasis comprehensive prediction model according to the key image features and the key clinical features by adopting a support vector machine algorithm. According to the embodiment, the model is established by adopting the random forest algorithm and the support vector machine algorithm, the prediction model can be established based on the structure risk minimum principle, and the problem of over-learning can be avoided, so that the constructed prediction model is more stable and accurate.

The invention relates to the field of road traffic safety, and discloses an intelligent sensing early warning method and system based on a vehicle state and a driving environment. The method comprisesthe steps of obtaining the state and driving environment parameters of a vehicle, carrying out data preprocessing on the state of the vehicle and the driving environment parameters, carrying out safety characteristic index quantification, analyzing the scalar indexes by using a method of combining single factor analysis and principal component analysis, and establishing a safety prediction modelbased on the vehicle state and the driving environment, and establishing an early warning strategy according to the safety prediction model based on the vehicle state and the driving environment. Thevehicle state information and the driving environment information are fused; a safety prediction model based on the vehicle state and the driving environment is constructed; the intelligent sensing early warning system based on the vehicle state and the driving environment is provided based on the safety prediction model, effective identification and pre-judgment of the vehicle operation safety are achieved, the early warning accuracy is high, the road traffic accident rate is reduced, and the road traffic safety is guaranteed.

The invention belongs to the technical field of medical treatment, and particularly discloses a system for identifying causes of fever of unknown origin of children. The system comprises a data acquisition module which is used for acquiring demographic index data and examination item index data of a sample; a data processing module used for receiving the data information acquired by the data acquisition module, preprocessing the data and then performing single-factor analysis on the preprocessed data; a risk factor acquisition module used for receiving the data subjected to single-factor analysis, performing regression analysis on the data and screening risk factors in the data; and an identification module internally provided with a neural network model, wherein the neural network model receives and analyzes the risk factor data and identifies the diagnosis direction of the causes of fever in children. According to the technical scheme, through cooperation of all the modules, collection and screening of sample information are achieved, and the diagnosis direction of causes of fever in children is identified according to the collected information.

The invention relates to a response surface method optimized subcritical water extraction method for ursolic acid in glossy privet leaves. The method comprises the following steps that a subcritical water method is used for extracting ursolic acid in the glossy privet leaves; a reaction still is cooled; solid-liquid separation is carried out; ethyl alcohol is added into sediment, shocking dissolution is carried out, and centrifugal separation is carried out; the content of ursolic acid in supernate is measured through high performance liquid chromatography (HPLC); through single factor analysis, the extraction process is optimized through a response surface optimizing method. Compared with the prior art, the method has the advantages of being simple and controllable in process, short in extraction time, clean, environmentally friendly, high in extraction rate, low in cost and the like, and is suitable for industrial production.

The invention provides a method for researching postoperative recurrence related factors of a non-small cell lung cancer patient based on multi-omics in a CT mode, and belongs to the field of computer-aided medicine. The method comprises the following steps: firstly, carrying out segmentation and image omics feature extraction on lung tumors; secondly, fusing the image omics characteristics with other omics characteristics to obtain a tumor multi-omics characteristic original data set; secondly, performing single-factor analysis on the multi-omics characteristic data and the recurrence condition of the patient by adopting an independent sample T test method to obtain recurrence related factors with statistical difference; and finally, substituting the factors with statistical difference into a Logistic regression analysis method to screen out significant characteristic factors related to recurrence of the non-small cell lung cancer, and constructing a model for verification. The invention provides a new scheme for exploring the relationship between the multi-omics characteristics under the CT mode and the postoperative recurrence condition of the patient.

The invention provides a soil erosion evaluation system, which mainly comprises a soil erosion single factor analysis module and a soil erosion comprehensive analysis module, and is characterized in that the soil erosion single factor analysis module is used for analyzing a corresponding relationship between each erosion factor and a soil erosion condition; wherein the erosion factors mainly comprise a rainfall erosion force factor, a soil erodibility factor, a terrain factor, a vegetation coverage factor and a water and soil conservation measure factor; and the soil erosion comprehensive analysis module is used for analyzing the weight of each erosion factor and the spatial and temporal distribution of erosion force. According to the method, the soil erosion is evaluated from the multi-dimensional perspective of arbitrary combination of multiple elements influencing the soil erosion, the method can better serve soil erosion mechanism research, and a reference can also be provided forregional resource environment evaluation.

The invention discloses a method for establishing a suicide risk prediction model of a tumor patient. The method comprises the following steps: collecting demographic data; collecting disease-relateddata of a tumor patient; collecting clinical related characteristic data of the tumor patient; collecting the anxiety and depression level, social support conditions, life quality, family functions and smegmatis conditions of the patient in a tumor rehabilitation period; taking whether a tumor patient has a suicide idea or not as a dependent variable to carry out single-factor analysis on relatedfactors possibly influencing the suicide idea, and establishing a column diagram prediction model; and carrying out distinguishing degree and calibration degree evaluation on the column diagram prediction model. The calibration degree of the prediction model can be evaluated through goodness-of-fit test and a correction curve, and the sensitivity and specificity are adopted to determine the prediction truncation value of the column diagram prediction model. According to the invention, the suicide risk grade of a tumor patient can be accurately predicted; the suicide risk of a tumor patient canbe evaluated in time at the terminal, and the clinical decision-making efficiency is improved.

The invention discloses a main effect factor analysis method and equipment based on a causal model. The main effect factor analysis method based on the causal model comprises the following steps: constructing a Bayesian network model, and obtaining direct influence factors of a result node; obtaining an influence path set of influence of each direct influence factor on a result node; when the influence path set comprises indirect influence paths, analyzing the influence degree of the direct influence factors on the result nodes jointly based on the node interaction effect, and otherwise, calculating the influence degree of the direct influence factors on the result nodes directly by adopting an independent influence degree analysis method; and selecting the direct influence factor corresponding to the maximum value from all the influence degrees as a main effect factor. According to the method, the problem of factor isolation of a single-factor analysis method is solved, the main effect factor is extracted on the basis of fully analyzing the linkage interaction influence between the factors, and the reasonability and credibility of the main effect factor analysis conclusion are improved.

The invention discloses a method for optimizing a supercritical CO2 extraction process of agilawood essential oil. The method comprises the following steps: S1, pretreatment: obtaining dry agilawood powder; S2, single-factor analysis: carrying out particle size analysis, extraction pressure analysis, extraction temperature analysis and extraction time analysis; and S3, response surface analysis: selecting the first three factors of granularity, extraction pressure, extraction temperature and extraction time contribution degree as reference factors, and taking the extraction rate of the agilawood essential oil as an evaluation index to perform response surface analysis. The contribution degree of each factor to the extraction rate of the agilawood essential oil is obtained through single factor analysis. A mathematical model is established by utilizing a response surface analysis method. Key factors and interactions thereof are selected by utilizing a response surface diagram of the model: the granularity is 30-50 meshes, the extraction temperature is 40 DEG C, the extraction pressure is 18MPa, the extraction time is 2h, the extraction rate of the agilawood essential oil reaches 0.688%, and the reliability of a result is verified by GC-MS (Gas Chromatography Mass Spectrometry), so that the method is an optimization method suitable for a supercritical CO2 fluid extraction processfor agilawood essential oil.

The invention discloses a type 2 diabetes prediction method and system based on the human body composition. The method comprises the following steps: acquiring basic information of a patient and related data of the human body composition; taking whether the follow-up visited patient suffers from type 2 diabetes or not as a dependent variable, taking the obtained factors as independent variables to carry out single-factor analysis, and screening factors with statistical difference according to the P value of each factor; carrying out multi-factor Logistic regression analysis on independent variables with statistical differences in a single-factor analysis result, selecting a model with the best fitting degree according to AIC values, determining a regression coefficient which is finally incorporated into the independent variables, and representing the regression coefficient by adopting a Nomogram column graph; and predicting the onset risk of the type 2 diabetes of the patient according to the predicted cut-off value of the column graph.

The invention belongs to the field of biotechnology, and particularly discloses an AWIA scoring model and a building method thereof. The building method is characterized by comprising the following steps: screening risk factors with difference between two groups through single factor analysis first; further selecting variables through a stepwise regression method to be taken as independent variables of multivariate Logistic analysis finally; obtaining a regression coefficient Beta of each independent variable after the multivariate Logistic analysis; adopting the lowest Beta value as a radix number to calculate multiples of the Beta values of the other independent variables to the independent variables, which are the corresponding scores of the independent variables; scoring a case group and a control group according to the scoring model; confirming critical scores with relatively high predictive value by drawing an ROC curve of a subject. After adoption of the model, diagnosis, treatment and rescue success rates of the AWIA are improved remarkably; the AWIA scoring model is further subjected to clinical practice and popularization.

The invention discloses a method for establishing a prognosis model for treating advanced liver cancer patients by HAIC, which comprises the following steps: (1) acquiring sample data, and screening the sample data; (2) carrying out single-factor analysis and multi-factor analysis by adopting a Cox proportional danger model, and selecting factors influencing prognosis from the sample data; (3) establishing a column graph prediction model for predicting prognosis according to the sample data and the factors influencing prognosis; and (4) verifying the column graph prediction model by adopting c-index and a correction curve. The model established by the method can evaluate the hepatic artery perfusion chemotherapy prognosis model of the local advanced liver cancer patient, and the model is displayed in the form of a column diagram, so that the model is intuitive and simple, and is convenient for clinical doctors and patients to operate and carry out individualized prediction on the patients.

The present invention discloses an analytical method of desulfurization efficiency influence factors in a flue gas desulphurization system. The analytical method is based on an analytical method of a response surface, only by several groups of experiment data. Single factor analysis and multifactor interaction analysis can be performed, thereby shortening the analytical time, increasing the analytical efficiency, and providing a fine-quality combination of desulfurization efficiency influence factors under a condition of achieving an emission standard reasonably and effectively. The analytical method is convenient and rapid. The analytical method provided by the present invention is applicable to a series of problems of desulfurization efficiency influence factors analysis in a flue gas desulphurization system, has a very good practice guide value and significance on regulating desulfurization efficiency for an actual desulfurization system.

The invention relates to an early warning device for kidney non-evident injuries. The device comprises a liquid flow pipeline, foam detectors, a central processing unit, a microflow control driving probe, a biological matrix single factor analysis module and a power source, wherein the foam detectors are arranged at the two ends of the liquid flow pipeline, the central processing unit is connected with the foam detectors, the microflow control driving device is arranged on the liquid flow pipeline and electrically connected with the central processing unit, the biological matrix single factor analysis module is in data connection with the central processing unit and is further connected with the microflow control driving probe through a pipeline, the biological matrix single factor analysis module comprises a urine specific gravity detection electrode, a beta-2-microglobulin detection electrode and a cystatin C detection electrode, the power source is connected with the central processing unit, and the central processing unit is further provided with a wireless transmission port in an integrated mode. According to the early warning device, urine can be sucked in a urine collection bag system automatically through the microflow control driving probe, detection is performed according to a preset frequency, measurement does not need to be interrupted, the detection cost is low, the informationization degree is high, convenience and rapidness are achieved, and the development potential is huge.

The invention discloses a batched accurate medication decision support system, which comprises a batched data import module, a data feature generation module, a data analysis module, an intelligent retrieval and import module and an analysis result generation module. The batched data import module can import JSON data and Excel data, the data feature generation module can add corresponding feature variables according to different disease categories and generate a corresponding disease category data feature extraction table, and the data analysis module realizes exploratory analysis through a self-defined research queue and arbitrary selection of parameters. According to the invention, being capable of importing desensitized structured data in batches is taken as an entry point, risk prediction and drug recommendation of multiple regions can be completed through one-time use requests, a model is re-optimized through single-factor analysis and multi-factor analysis, drug risk prediction with times is provided for clinic, and finally data support is provided for formulation of clinical drug guide or expert consensus.

The invention provides a calculation method for prompting the Alzheimer's disease (AD) risk of old men by androgen. The method comprises the following steps: constructing an equation model, fitting anROC curve, and calculating a diagnosis boundary value cutoff value, so as to obtain an androgen laboratory reference value of amnetic mild cognitive impairment (aMCI). The method comprises the following specific steps: referring to a result of single-factor analysis of aMCI and cognitive normality comparison, selecting variables with statistical significance to further perform multi-factor logistic regression analysis to obtain independent prediction factors of aMCI, and constructing an equation; the method has a high prediction value, and can prompt possible aMCI risks.

The invention discloses an extraction process optimization method and a rapid extraction method of poison mushroom alpha-amatoxin. An ultrasonic extraction method is adopted to extract alpha-amatoxin, firstly, five test factors are screened through a single factor analysis test, then, a multi-factor multi-level test is designed through a response surface center combination method, the content of alpha-amatoxin is determined through an HPLC method, the extraction yield of alpha-amatoxin is calculated, optimal extraction process conditions of poison mushroom alpha-amatoxin are determined by integrating all experiment results, and then experiments are carried out for verification, so that the rapid and efficient ultrasonic extraction method of the poison mushroom alpha-amatoxin is finally obtained. The method is short in extraction time and high in efficiency.

In the screening process of the preparation method of the large radix bupleuri decoction, the Q-marker concept is introduced for the first time, and the large radix bupleuri decoction is screened according to the Q-marker screening result; main active substances, such as saikoside B2, baicalin, baicalein, wogonoside, naringin, neohesperidin, hesperidin and paeoniflorin, which are related to the drug effect of the big bupleurum decoction are taken as main efficacy markers for screening the preparation method of the big bupleurum decoction, and the improved preparation method of the big bupleurum decoction is screened through single factor analysis and orthogonal design. The large bupleurum decoction prepared by the method has obviously improved effects of relieving Shaoyang and purging heat accumulation.

The invention provides a deep shaft engineering water inrush disaster multi-source information evaluation method, and belongs to the technical field of mine water inrush risk evaluation. The method comprises the following steps: firstly, establishing a multivariate information evaluation system, then carrying out dimensionless treatment on original indexes, then calculating the weight of influencefactors, and finally constructing a water inrush risk evaluation model. According to the method, a water inrush risk evaluation index system is established according to factors influencing the vertical shaft water inrush risk, test methods of different influence factors are provided, then an analytic hierarchy process is used for determining weights of different factors, and finally the verticalshaft water inrush risk is evaluated based on a multi-source information set model, wherein the tectonic fissure zone, the fault fracture zone, the water-rich property, the water inrush point, the water conductivity and the crustal stress are selected to establish a multivariate information evaluation system. According to the method, accidental errors in the single factor analysis process are solved, a more comprehensive and accurate analysis method is provided, a reliable basis is provided for construction decision making, and the practicability is higher.