Font Size: a A A

Correlation Between The Eye Collateral Channels And TCM Symptom Elements Of Elderly Mild Cognitive Impairment And The Study Of A Discriminant Model

Posted on:2024-02-28Degree:MasterType:Thesis
Country:ChinaCandidate:L CaoFull Text:PDF
GTID:2544307097952589Subject:Medicine Rehabilitation learn
Abstract/Summary:PDF Full Text Request
Objectives:The aim of this study was to investigate the characteristics of the eye collateral channels and the distribution characteristics of the TCM symptom elements of the MCI population,as well as the internal relationship between them,in order to provide an objective basis for the early identification of MCI.Methods:This was a case-control study,in which the subjects were recruited from the health management center of the Second People’s Hospital of Fujian Province and community-dwelling elderly people in Fuzhou,and the eligible subjects who met the inclusion and exclusion criteria were divided into MCI and cognitively normal groups according to the 1:1 propensity matching method for age,gender,and years of education.We could collect the basic demographic information,the cognitive test results,the TCM symptom elements and the eye collateral channels of the subjects by on-site surveys using the general demographic information form,the neuropsychological test scale,the TCM symptom identification system and the Boao Eye Diagnostic Instrument.The two independent non-parametric test and the chi-square test were used to analyze the differences between the MCI group and the normal cognitive group in terms of the characteristics of eye collateral channels and TCM symptom elements;the frequency analysis and the principal component analysis were used to explore the distribution characteristics of TCM symptom elements in MCI groups;Using the Spearman correlation analysis to investigate the correlation between the common TCM symptom elements and the eye collateral channels of the MCI.The data of 316 cases were divided into 80% training set and 20% validation set,and the support vector machine,decision tree,neural network and random forest algorithms were applied to construct different discriminative models for mild cognitive impairment,in which using the TCM symptom elements and the eye collateral channels as independent variables and whether being MCI was the dependent variable.We selected the optimal model by comparing the model performance,in order to achieve early clinical identification of the MCI.Results:1.Of the 316 subjects,158 were in the MCI group and 158 were in the cognitively normal group.No statistical differences were seen between the two groups in terms of age,gender,and years of education.There were 85 females in their own group,accounting for 53.2% of the total numbers in each group;the median age was 66 years in both groups;the median years of education respectively were 9 and 11 years in the MCI group and the cognitively normal group.2.The results of the comparison of eye collateral channels between the two groups: the important eye collateral channels in the MCI group were red dots,dull brown spots,dull yellow fog diffusion,mounds,red blood veins,and tortuous blood veins,and their distribution frequencies were higher than those in the normal cognitive group,but the frequencies of dull pink spots were lower than those in the normal cognitive group,The frequency of "E","H&J" and "I" areas were higher than those of the normal cognitive group,which corresponded to "Heart","Kidney" and "Bladder" respectively.3.Distribution of the Chinese medical symptom elements in MCI: the common TCM disease location evidence elements in MCI group were Liver(98.1%),Kidney(96.8%),Tendons and bones(82.3%),Spleen(65.8%),Heart(55.7%),and Meridians(53.2%),and their scores of central,liver,and spleen evidence elements were higher than those of the cognitive normal group;the common TCM pathogenic evidence elements in MCI group were phlegm(94.3%),dampness(94.3%),and blood stasis(88.6%),The MCI group had higher scores for Phlegm(94.3%),Dampness(94.3%),Blood stasis(88.6%),Qi stagnation(84.8%),Heat(74.1%),Yang hyperactivity(67.7%),and Cold(51.9%),with higher scores for Phlegm,Dampness,Blood stasis,Qi stagnation,Heat,and Yang hyperactivity than the normal cognitive group;the MCI group had higher scores for Yin deficiency(97.5%),Blood deficiency(94.9%),Qi deficiency(90.5%),Yang deficiency(76.6%),Essence deficiency(65.8%),and Jin deficiency(53.8%),with Yin deficiency,Qi deficiency,Qi deficiency,and Jin deficiency symptom elements scores higher than those of the cognitive normal group.4.The results of the factor analysis of the TCM symptom elements showed that the evidence information of MCI group was clustered into 4 categories,in which the first category constituted deficiency of Qi and Blood syndrome,the second category constituted Hyperactivity Liver Yang syndrome,the third category constituted Damp and Blood stasis channels syndrome,and the fourth category constituted deficiency of Kidney vital fluid essences syndrome.5.The results showed that red dull pigmentation,red dull haze,yellow mound,red blood vein,and liver area were positively correlated with Hyperactivity Liver Yang syndrome(P < 0.05,P < 0.01),and dull yellow moon halo and light blood vein were negatively correlated with Hyperactivity Liver Yang syndrome(P < 0.05);pale mound and dull blood vein were positively correlated with Damp and Blood stasis channels syndrome(P < 0.05,P < 0.01),red dull spots and pink blood vessels were negatively correlated with Damp and Blood stasis channels syndrome(P < 0.05,P < 0.01);dull moon halo was positively correlated with deficiency of Kidney vital fluid essences syndrome(P < 0.05,P < 0.01),pink dull haze and red blood vessels were negatively correlated with deficiency of Kidney vital fluid essences syndrome(P < 0.05);small intestine area was negatively correlated with deficiency of Qi and Blood syndrome(P < 0.05).6.In this study,the support vector machine model was selected as the optimal discrimination model,with an accuracy of 71.85%,sensitivity of 0.857,specificity of 0.642,and AUC of 0.777.The overall performance was good.For discriminating MCI,the largest contribution among the eye collateral channels was dull yellow haze diffusion,followed by dull brown spots,and the largest contribution among the evidence elements was the pathological elements phlegm and qi stagnation,followed by blood stasis and yin deficiency.Conclusion:1.The eye collateral channels had some reference significance in identifying the TCM symptom elements,including red dots,dull brown spots,dull yellow mist diffusion,mounds,and tortuous blood veins,and they played important roles in the TCM symptom diagnosis of the MCI.2.The main characteristics of the eye collateral channels of MCI involving to blood stasis,damp-heat,phlegm and other pathological elements,and its locational features were concentrated in the "Heart","Kidney and Bladder",suggesting that the core pathological sites were in the Heart and Kidney,which reflected the pathological characteristics of the deficiency of MCI in Traditional Chinese Medicine.3.The MCI discriminative model based on the characteristics of the eye collateral channels and the TCM symptom elements,combined with the support vector machine algorithm,had great feasibility and accuracy,and provided a reliable basis for the early identification of MCI.
Keywords/Search Tags:Mild Cognitive Impairment, Traditional Chinese Medicine Certificate, Eyes Collateral Channels, Five rings and eight octants, Machine Learning
PDF Full Text Request
Related items