CN1891144A - Stroke pre-warning detector - Google Patents
Stroke pre-warning detector Download PDFInfo
- Publication number
- CN1891144A CN1891144A CN 200510027436 CN200510027436A CN1891144A CN 1891144 A CN1891144 A CN 1891144A CN 200510027436 CN200510027436 CN 200510027436 CN 200510027436 A CN200510027436 A CN 200510027436A CN 1891144 A CN1891144 A CN 1891144A
- Authority
- CN
- China
- Prior art keywords
- apoplexy
- regression
- main constituent
- block
- principal component
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Abstract
The present invention discloses an apoplexy warning detection apparatus. It is aimed at utilizing existent cerebrovascular hemodynamics to detect parameter information and provide an accurate and practical apoplexy warning measure. It includes the structure including cerebrovascular hemodynamics delection apparatus, and is characterized by that said invention also includes the following several modules: main component analysis module, logistic regression module and apoplexy pathogenic probability estimation module. Said invention also provides the concrete action and function of the above-mentioned every module, and also provides the working principle of said apoplexy warning detection apparatus and its conrete operation method.
Description
Technical field
The present invention relates to medical instruments field, be specifically related to a kind of apoplexy pre-warning detector that makes up with cerebrovascular blood kinetic measurement parameter and risk of stroke factor.
Background technology
Apoplexy has risen to first of the total cause of death of China male, and in a single day apoplexy takes place, and about 3/4 patient leaves over deformity in various degree.Therefore, should before apoplexy takes place, accurately predict the onset risk of apoplexy, effectively identify the high-risk individuality of apoplexy, take effective measures and carry out early prevention and treatment, could reduce the M ﹠ M of apoplexy.
In apoplexy early warning field long term studies and exploration were arranged both at home and abroad, its main method is to detect index by the investigation of risk of stroke factor and clinical laboratory, sets up apoplexy early warning means.But These parameters can not accurately reflect the injured degree of cerebrovascular, also has individual susceptibility difference and deficiencies such as used lab index sensitivity and specificity difference, and the accuracy of early warning means is difficult to reach ideal level always.Therefore, still lack at present and can be applied to clinical and community prevention, method for early warning accurately.
Summary of the invention
The objective of the invention is to utilize existing cerebrovascular hemodynamics detected parameters information, set up a kind of detector that is used for the apoplexy early warning, a kind of accurate, practical, feasible apoplexy early warning means are provided.
Apoplexy pre-warning detector of the present invention, the structure that comprises the cerebrovascular hemodynamics detector, it is characterized in that: by the blood motion mathematic(al) parameter and the hemodynamic parameter of cerebrovascular hemodynamics detector output, together with census of population, detection database data, import the principal component analysis module together and carry out statistical analysiss such as main constituent, export each parameter main constituent factor contributions; Export main constituent factor contributions and risk of stroke factor information to the Logistic regression block, Logistic regression block output regression model is to apoplexy probability estimation block then; Export the apoplexy incidence rate by apoplexy incidence rate estimation block at last, and can feed back to the cerebrovascular hemodynamics detector.
Implement the obvious advantage after this technology: the present invention is directed to the apoplexy early warning and carried out long-term, deep scientific research.Studies confirm that, before apoplexy takes place, the regular ANOMALOUS VARIATIONS of cerebrovascular hemodynamics detected parameters promptly occurs; The apoplexy morbidity is closely related unusually with the cerebrovascular hemodynamics detected parameters; On the basis of large sample census of population and detection, by data analysis, set up apoplexy Early-warning Model and detector, can assess the incidence rate of apoplexy quantitatively, exactly.Above-mentioned discovery is extremely important for early screening and identification high-risk individuality of apoplexy and stroke prevention.
Description of drawings
Fig. 1 is an apoplexy pre-warning detector structural representation:
Among the figure, 1-cerebrovascular hemodynamics detector; 2-is the blood motion index; 3-hematodinamics index; 4-2.5 ten thousand apoplexy formation census of population, detection data base; 5-principal component analysis module; 6-main constituent factor contributions; 7-risk of stroke factor investigation information; The 8-Logistic regression block; The 9-Logistic regression model; 10-apoplexy incidence rate estimation block; 10-apoplexy incidence rate.
The specific embodiment
Below in conjunction with accompanying drawing the present invention is further described:
See Fig. 1, by cerebrovascular hemodynamics detector one group of blood motion mathematic(al) parameter 2 of 1 output and hemodynamic parameter 3, together with census of population, detection database data 4, import principal component analysis module 5 together and carry out statistical analysiss such as main constituent, export the main constituent factor contributions 6 of each parameter 2,3; Export main constituent factor contributions 6 and risk of stroke factor information 7 to Logistic regression block 8, through regression treatment, Logistic regression block 8 output regression models 9 are to apoplexy probability estimation block 10; Estimate by apoplexy probability estimation block 10 at last, and output apoplexy incidence rate 11.The apoplexy probability of happening can be exported by the output system of hemodynamic examination instrument, and prints the result.
See Fig. 1, one group of blood motion mathematic(al) parameter 2 of cerebrovascular hemodynamics detector 1 output and hemodynamic parameter 3, and 2.5 ten thousand census of population, detect database data 4, input to principal component analysis module 5 jointly.This module with the 2.5 ten thousand routine apoplexy crowds' of cohort study investigation, detect and the data of following up a case by regular visits to is a master data, cerebrovascular hemodynamics is detected data carry out principal component analysis, obtain preceding four main constituents of these data, its accumulation contribution rate reaches 94.6%, thereby obtain the factor contributions 6 of each each main constituent of parameter, and export Logistic regression block 8 to.
In Logistic regression block 8, whether generation with apoplexy in the process of following up a case by regular visits to is dependent variable, preceding four main constituents, risk of stroke factor 7 (containing parameters such as age, sex, hypertension, heart disease, diabetic history, Body Mass Index) that the factor contributions 6 that obtains with principal component analysis comprises are independent variable, import the Logistic regression model and carry out regression analysis, obtain entering the variable of regression equation: first, second, third, fourth main constituent and hypertension history, age and sex, regression coefficient in conjunction with selected variable obtains regression model 9.
Apoplexy probability estimation block 10, set up equation according to regression model 9: apoplexy probability=1.025* first principal component-0.136* Second principal component,-0.299* the 3rd main constituent+0.033* the 4th main constituent+0.543* hypertension history-0.681* sex+0.035* age+constant term, and equation estimation apoplexy probability 11 in view of the above.At this moment, can be with apoplexy incidence rate and together output print result of hemodynamic parameter.
Also can be with the following method, estimation apoplexy early warning effectiveness parameters: calculate the P of each individual apoplexy among the formation crowd, again according to probability and actual incidence, draw the ROC curve, area under curve is 0.855.Follow up a case by regular visits to the data in 3 years with the formation crowd, get different point of cut-offs according to the size of prediction probability, when value 0.05, the exponential value of Youden is maximum, and 0.05 is the best intercept point of early warning.
With 0.05 as point of cut-off, all experimenters are divided into four kinds of situations such as true positives, false positive, true negative, false negative, and import four fold table (table 1).
The prediction effect analysis of table 1. apoplexy forecast model
Model prediction | Follow-up results | Add up to | |
The morbidity number | The number of not falling ill | ||
Positive | 130 | 5207 | 5337 |
Negative | 31 | 19042 | 19073 |
Add up to | 161 | 24249 | 24410 |
According to table 1, can be calculated as follows apoplexy early warning effectiveness parameters:
Claims (2)
1, a kind of apoplexy pre-warning detector of the present invention, the structure that comprises inspection and measuring instrument for cerebrovascular function 1, it is characterized in that: by cerebrovascular hemodynamics detector one group of blood motion mathematic(al) parameter 2 of 1 output and hemodynamic parameter 3, together with census of population, detection database data 4, import principal component analysis module 5 together and carry out statistical analysiss such as main constituent, export the main constituent factor contributions 6 of each parameter 2,3; Export main constituent factor contributions 6 and risk of stroke factor information 7 to Logistic regression block 8, through regression treatment, Logistic regression block 8 output regression models 9 are to apoplexy probability estimation block 10; Estimate by apoplexy probability estimation block 10 at last, and output apoplexy incidence rate 11.The apoplexy probability of happening can be exported by the output system of hemodynamic examination instrument, and prints the result.
2, according to the described a kind of apoplexy pre-warning detector of claim 1, it is characterized in that: in Logistic regression block 8, whether generation with apoplexy in the process of following up a case by regular visits to is dependent variable, preceding four main constituents that the factor contributions 6 that obtains with principal component analysis comprises, risk of stroke factor 7 (contains the age, sex, hypertension, heart disease, diabetic history, parameters such as Body Mass Index) be independent variable, import the Logistic regression model and carry out regression analysis, obtain entering the variable of regression equation: first, second, the 3rd, the 4th main constituent and hypertension history, age and sex, regression coefficient in conjunction with selected variable obtains regression model 9; Apoplexy probability estimation block 10, set up equation according to regression model 9: apoplexy probability=1.025* first principal component-0.136* Second principal component,-0.299* the 3rd main constituent+0.033* the 4th main constituent+0.543* hypertension history-0.681* sex+0.035* age+constant term, and equation estimation apoplexy probability 11 in view of the above.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN 200510027436 CN1891144A (en) | 2005-07-01 | 2005-07-01 | Stroke pre-warning detector |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN 200510027436 CN1891144A (en) | 2005-07-01 | 2005-07-01 | Stroke pre-warning detector |
Publications (1)
Publication Number | Publication Date |
---|---|
CN1891144A true CN1891144A (en) | 2007-01-10 |
Family
ID=37596533
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN 200510027436 Pending CN1891144A (en) | 2005-07-01 | 2005-07-01 | Stroke pre-warning detector |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN1891144A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104055524A (en) * | 2014-06-30 | 2014-09-24 | 山东大学 | Brain function connection detection method and system based on near infrared spectrum |
CN109599175A (en) * | 2018-12-04 | 2019-04-09 | 中山大学孙逸仙纪念医院 | A kind of analysis of joint destroys the device and method of progress probability |
CN110022767A (en) * | 2016-12-02 | 2019-07-16 | 心脏起搏器股份公司 | Row apoplexy is poured into using blood pressure peak to detect |
CN110612057A (en) * | 2017-06-07 | 2019-12-24 | 柯惠有限合伙公司 | System and method for detecting stroke |
CN113257421A (en) * | 2021-05-31 | 2021-08-13 | 吾征智能技术(北京)有限公司 | Method and system for constructing hypertension prediction model |
-
2005
- 2005-07-01 CN CN 200510027436 patent/CN1891144A/en active Pending
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104055524A (en) * | 2014-06-30 | 2014-09-24 | 山东大学 | Brain function connection detection method and system based on near infrared spectrum |
CN104055524B (en) * | 2014-06-30 | 2016-05-04 | 山东大学 | Brain function based near infrared spectrum connects detection method and system |
CN110022767A (en) * | 2016-12-02 | 2019-07-16 | 心脏起搏器股份公司 | Row apoplexy is poured into using blood pressure peak to detect |
CN110612057A (en) * | 2017-06-07 | 2019-12-24 | 柯惠有限合伙公司 | System and method for detecting stroke |
CN110612057B (en) * | 2017-06-07 | 2022-05-31 | 柯惠有限合伙公司 | System and method for detecting stroke |
CN109599175A (en) * | 2018-12-04 | 2019-04-09 | 中山大学孙逸仙纪念医院 | A kind of analysis of joint destroys the device and method of progress probability |
CN113257421A (en) * | 2021-05-31 | 2021-08-13 | 吾征智能技术(北京)有限公司 | Method and system for constructing hypertension prediction model |
CN113257421B (en) * | 2021-05-31 | 2023-09-15 | 吾征智能技术(北京)有限公司 | Construction method and system of hypertension prediction model |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Quer et al. | Wearable sensor data and self-reported symptoms for COVID-19 detection | |
Shang et al. | Systematic review and meta-analysis of flow cytometry in urinary tract infection screening | |
Hou et al. | Counting citations in texts rather than reference lists to improve the accuracy of assessing scientific contribution: Citation frequency of individual articles in other papers more fairly measures their scientific contribution than mere presence in reference lists | |
Xiang et al. | Periodontitis‐specific molecular signatures in gingival crevicular fluid | |
CN101981446A (en) | Method and system for analysis of flow cytometry data using support vector machines | |
CN1891144A (en) | Stroke pre-warning detector | |
CN112786203A (en) | Machine learning diabetic retinopathy morbidity risk prediction method and application | |
Fusaro et al. | Marker‐dependent hazard estimation: An application to AIDS | |
Ma | Discovering association with copula entropy | |
CN103761451B (en) | Biomarker combined recognising method and system based on biomedical big data | |
JP5807956B2 (en) | Method for measuring crop growth frequency of soil and method for judging fake production | |
CN203303031U (en) | Diabetes non-invasive detecting device based on glycation end product fluorescence spectrum | |
CN111175480A (en) | Method for calculating gender and age by blood biochemical indexes | |
RU2406078C2 (en) | Method for detection and identification of biological microobjects and their nanocomponents and related device for implementation thereof | |
Banks et al. | A division-dependent compartmental model with cyton and intracellular label dynamics | |
Chen et al. | Application of PLSR in rapid detection of glucose in sheep serum | |
S Subramani et al. | Classification learning assisted biosensor data analysis for preemptive plant disease detection | |
Faulkner et al. | The role of epigenetic and biological biomarkers in the diagnosis of periodontal disease: a systematic review approach | |
CN106124763A (en) | For diagnosing the biomarker combinations thing of cervicitis and cervical cancer | |
CN112716447A (en) | Oral cancer classification system based on deep learning of Raman detection spectral data | |
TWM651915U (en) | A system that uses the degree of mitochondrial aging to interpret physiological information of tested organisms | |
Nový | Saliva and biofilm-based diagnostics: a critical review of the literature concerning sialochemistry | |
CN110070942A (en) | A kind of chronic liver disease risk evaluating system promoting tree-model based on gradient | |
TWM650258U (en) | System using telomere length detection to determine true age of tested organism | |
US20230019703A1 (en) | Personal health tracking urinalysis device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
WD01 | Invention patent application deemed withdrawn after publication |