Eye Collateral Channel Characteristic Analysis and Identification Model Construction of Mild Cognitive Impairment

Abstract

ObjectiveTo investigate the eye collateral channel characteristics of mild cognitive impairment (MCI) population, and to build an MCI identification model based on machine learning algorithms to provide an objective basis for early recognition of MCI.MethodsA total of 316 subjects from 5 communities in Fuzhou City, Fujian Province and the Health Management Center of the Second People's Hospital Affiliated to Fujian University of Traditional Chinese Medicine were recruited from April to December 2022. Eligible subjects were matched on a 1∶1 propensity score according to sex, age, and years of education and were divided into MCI group and normal cognition group, with 158 cases in each group. Using the general demographic data sheet, neuro-psychological test scale, syndrome identification system of TCM and Boao visual diagnosis instrument, basic information, cognitive test results, and information on eye collateral channel characteristics and TCM symptom elements of the subjects were collected. Two independent non-parametric test and chi-square test were used to analyze the differences between the MCI group and the normal cognition group in terms of eye collateral channel characteristics and TCM symptom elements. Frequency analysis and principal component analysis were used to explore the distribution characteristics of TCM symptom elements in MCI patients. The study data of 316 cases were then randomly divided into 80% training set and 20% validation set. Different MCI identification models were constructed using support vector machine, decision tree, artificial neural network and random forest algorithm, with MCI eye collateral channel characteristics and TCM syndrome elements as independent variables and onset of MCI as a dependent variable. By comparing the model performance, the optimal model was selected to achieve early clinical recognition.ResultsThe important eye collateral channel characteristics of the MCI group were red dots, dull brown spots, dull yellow fog diffusion, mounds, red blood veins, and tortuous blood veins (P<0.05), their distribution frequency was higher than that of the normal cognition group (P<0.05), but the distribution frequency of dull pink spots was lower than that of the normal cognition group (P<0.05). The heart, spleen, and liver scores of the MCI group were higher than those of the normal cognition group (P<0.05). Comparison of TCM substantive syndrome element scores between two groups, the scores of phlegm, dampness, blood stasis, Qi stagnation, heat, yang hyperactivity, heatstroke, and food accumulation in the MCI group were all higher than those in the normal cognition group (P<0.05). Comparison of TCM deficiency syndrome element scores between two groups, the scores of yin deficiency, blood deficiency, Qi deficiency, fluid deficiency, and Qi depression in the MCI group were all higher than those in the normal cognition group (P<0.05). The common pathological evidence elements in the MCI group were liver, kidney, muscles and bones, spleen, heart, and meridians, and the syndrome element scores of the heart, liver, and spleen were higher than those in the normal cognition group (P<0.05). Common disease location syndrome elements in the MCI group are liver, kidney, muscles and bones, spleen, heart, and meridians. Common real syndrome elements are phlegm, dampness, blood stasis, and Qi stagnation. Common deficiency syndrome elements are yin deficiency, blood deficiency, Qi deficiency, and yang deficiency. Based on the MCI identification model of eye collateral channel characteristics and TCM syndrome elements, the support vector machine model was selected as the optimal discriminant model, with an accuracy of 73.08%, a sensitivity of 0.677, a specificity of 0.765, and an AUC of 0.777. The overall performance was relatively good.ConclusionCompared with normal cognitive subjects, MCI patients have unique eye collateral channel characteristics. The MCI identification model constructed based on the eye collateral channel characteristics and TCM syndrome elements, combined with the support vector machine algorithm, has good feasibility and accuracy, which can provide a basis for the objective recognition of MCI.

Keywords

• mild cognitive impairment;eyes collateral channels;machine learning;support vector machine