Research On The Nomogram Prediction Model For Stroke Risk Based On "observing The Eyes To Identify The Disease

Purpose:Study the differences in the characteristics of the white eye meridians between stroke patients and normal individuals,analyze the risk factors of stroke onset,and construct a Nomogram prediction model for stroke onset risk.Material and method:Using the method of prospective cohort study,a natural dynamic cohort was established for one year from January 2022 to December 2022.The subjects were stroke patients admitted to the Second Department of Brain Rehabilitation of the Affiliated Hospital of Liaoning University of Traditional Chinese Medicine and residents of nearby communities.A total of 123 subjects were included,including 50 patients diagnosed as stroke recovery and 73 community residents.Collect the general data of the subjects,observe the white eye meridian and conduct semi-quantitative analysis of the conjunctival microcirculation of the subjects,use SPSS26.0 software for statistical processing,study the difference of the position,shape,color and other characteristics of the abnormal white eye meridian between the two groups of people,and use single factor analysis and multiple factor logistic regression analysis to determine the independent risk factors of stroke,The Nomogram prediction model of stroke incidence risk was built by using R stidio software,and the correction curve was used for internal validation of the model,and the decision curve was used to evaluate the prediction efficiency of the model.Results:1 Compared with the general data of the two groups,there was no significant difference in sex,height and weight between the stroke cohort and the normal cohort(P>0.05).There was significant difference in age(P<0.05).2 Compare the differences in the number,position,shape and color of abnormal white eye collaterals between the two groups.The average number of abnormal white eye collaterals among the subjects in the stroke cohort was large.The location of the abnormal white eye collaterals was mostly in the hepatobiliary area and the upper focus area.The morphology was mostly fuzzy,with many bifurcations,and the color was mostly dark red or light.There were no significant statistical characteristics in the characteristics of the white eye meridians of the normal subjects.3 Compared with the semi-quantitative analysis data of bulbar conjunctival microcirculation in the two groups,there was no significant difference between the stroke cohort and the normal cohort in terms of the diameter of the arterioles(50X),the diameter of the arterioles(50X),the diameter ratio of the arterioles(V/A),the edge irregularity(20X total field of vision),saccular expansion(20X total field of vision),ischemic area(20X total field of vision),exudation or edema,and hemosiderosis(P>0.05).There were significant differences between the two groups in clarity,reticular structure,abnormal course,uneven thickness(20X total field of vision),microangioma(20X total field of vision),A-V short circuit branch and blood color(P<0.05)4 Lasso regression analysis showed that sex,age,clarity,reticular structure,A-V short circuit branch,uneven thickness,microangioma,exudation or edema,a total of 8 factors were independent risk factors affecting the incidence of stroke.5 The Nomogram prediction model of stroke incidence risk was established.After verification,the sensitivity of the model was 100.0%,the specificity was 94.0%,and the AUC was 0.978.The calibration curve fits well with the reference line P=0.971.Clinical applicability is good when the predictive value range is between 4% and 93%.Conclusion:1 Construct a Nomogram prediction model for stroke incidence risk.This model has demonstrated a certain degree of reliability through internal validation,and may easily and intuitively predict the risk of stroke among participants.2 The model includes the morphology and color characteristics of the white eye veins.There are 8 risk factors including subject gender,age,clarity of white eye veins,changes in network structure,uneven diameter of fine arteriovenous vessels,as well as the number of A-V short circuiting branches,microangiomas,exudative and edema lesions.3 Stroke patients have more abnormal white eye meridians than normal individuals,mostly distributed in the liver and gallbladder areas and the upper focal area.The abnormal morphology is characterized by blurry patches,more branching,and the color is often light or deep red.4 The white eye collateral vessels in stroke patients exhibit poor clarity,obvious network structure,and abnormal routing,with color changes to light red or deep red.These manifestations are related to the uneven thickness of arterial and venous diameter changes,and the formation of microvascular tumors and A-V short circuiting branches.5 The seven colors and eight forms of the white eye meridian observed and recorded in this study are consistent with those recorded in Professor Peng Jingshan’s book "Eye Acupuncture Therapy".