| ObjectiveThis topic aims to collect facial images of patients with Chronic Gastritis(CG)by using the self-developed TCM diagnostic instrument V1.0,analyzes the color and morphological parameters of facial skin,and adopts syndrome differentiation method and image analysis technology.This topic explores the facial skin color and characteristics of different types patients with CG,which is valuable and significant for the diagnosis of traditional Chinese medicine diagnosis.This topic so as to provides a reference for the study of facial diagnosis,and provides new methods for TCM clinical disease assist diagnosis and health assessment.Methods1.Clinical data collection and processing: From January 2022 to December 2022,238 CG patients who were visited by the department of spleen and stomach clinics of Fujian university of traditional chinese medicine,affiliated to Jinjiang hospital of traditional chinese medicine,were selected.Among them,184 patients with chronic non-atrophic gastritis(CNAG),54 patients with chronic atrophic gastritis(CAG),and 73 healthy volunteers as comparison.Then we collect the general information of patients,the four diagnostic methods of traditional Chinese medicine,the information of gastroscopy and pathological diagnosis,and determine pathological elements and syndrome through syndrome differentiation.2.Extraction of facial color and shape parameters: The facial images of all the subjects and the healthy group were collected.First,the YOLOv5 model was used to predict the region and category,then the Lab color model was used to extract the facial skin color parameters,the second-order gray scale statistics were used to measure the gray co-incidence matrix model to extract the facial skin texture parameters,and the first-order gray scale histogram was used to measure the facial skin wrinkles parameters.3.Correlation analysis between facial color and shape parameters and CG disease location and disease characteristics: SPSS 26.0 statistical software was used for statistical processing of the data,Pearson correlation analysis was used for data samples with normal distribution,and Spearman correlation analysis was used for data samples with non-normal distribution.The correlation between two variables is expressed by correlation coefficient.Results1.The distribution of syndrome elements:(1)The common disease syndrome of CNAG and CAG patients from high to low is:stomach,liver,and spleen.(2)The common syndromes of CNAG are qi stagnation,dampness,Yang deficiency,heat,phlegm,Yin deficiency,qi deficiency,food accumulation and blood stasis.The common syndromes of CAG are: dampness,qi stagnation,phlegm,heat,Yang deficiency,Yin deficiency,qi deficiency and food accumulation.2.Facial color parameters:(1)Color parameters: the left cheek L value,right cheek L value,nasal area L value,and mandibular area L value in the CNAG group were lower than those in the healthy group(P<0.01,P<0.05),and the frontal area a value,left cheek a value,right cheek a value,nasal area a value,and mandibular area a value were higher than those in the healthy group(P<0.01,P<0.05);the left cheek L value,nasal area L value,and mandibular area L value in the CAG group were lower than those in the healthy group(P<0.05),and the frontal area a value,left cheek a value,nasal area a value,mandibular area a value,frontal area b value,and right cheek b value were higher than those of the healthy group(P<0.05);the frontal area b value,left cheek b value,right cheek b value,and nasal area b value of the CNAG group were lower than those of the CAG group(P<0.01,P<0.05).(2)Textural parameters: frontal area ASM,left cheek ASM,right cheek ASM,nasal area ASM,mandibular area ASM,left cheek COR,mandibular area COR were significantly lower in the CNAG group than in the healthy group(P<0.01),and frontal area CON,left cheek CON,right cheek CON,mandibular area CON,left cheek ENT,right cheek ENT,nasal area ENT,mandibular area ENT were significantly higher than in the healthy group(P<0.01);frontal area ASM,left cheek ASM,right cheek ASM,nasal area ASM,mandibular area ASM,left cheek COR,mandibular area COR were significantly lower in the CAG group than in the healthy group(P<0.01),and frontal area CON,left cheek CON,right cheek CON,mandibular area CON,frontal area ENT,left cheek ENT,right cheek ENT,nasal area ENT,mandibular area ENT were significantly higher than in the healthy group(P<0.01);ASM in the mandibular region was higher in the CNAG group than in the CAG group(P0.05).(3)Wrinkle parameters: the mean value of nasal region,middle value of right cheek,and middle value of nasal region in the CNAG group were lower than those in the healthy group(P<0.01,P<0.05);the mean value of nasal region,middle value of nasal region,and middle value of mandibular region in the CAG group were lower than those in the healthy group(P<0.05),and the SD value of frontal region and SD value of nasal region were higher than those in the healthy group(P<0.01);the SD value of frontal region and SD value of nasal region in the CNAG group were lower than those in the CAG group(P<0.01).3.Comparison of facial color and shape parameters of pathological evidence: SD value of mandibular region was lower in the qi stagnation evidence group than in the non-qi stagnation evidence group(P<0.05);L value of right cheek and ENT of frontal region were lower in the wet evidence group than in the non-wet evidence group(P<0.05),and ASM of mandibular region was higher than in the non-wet evidence group(P<0.05);L value of left cheek,L value of right cheek,ASM of frontal region,ASM of left cheek,ASM of right cheek,ASM of nasal region,ASM of mandibular region were higher in the yang deficiency evidence group(P<0.01,P<0.05).ASM,left cheek mean,right cheek mean,left cheek middle value,right cheek middle value,and mandibular area middle value were higher than those of the non-Yang deficiency evidence group(P<0.01,P<0.05);frontal area a value,right cheek b value,frontal area CON,right cheek CON,nasal area CON,mandibular area CON,frontal area ENT,right cheek ENT,and nasal area SD values were lower than those of the non-Yang deficiency evidence group(P<0.01,P<0.05).4.Correlation of facial color and shape parameters with TCM evidence:(1)Correlation parameters with disease evidence elements: among the disease evidence elements,31 parameters with positive and negative correlation were obtained.There were 1parameter with correlation with the gastric evidence element,and the mandibular area CON(r=-0.164,P<0.05)was negatively correlated with the stomach;23 parameters with correlation with the spleen evidence element,the frontal area L value(r=0.231,P<0.05),the left cheek L value(r=0.213,P<0.05),the mandibular area L value(r=0.236,P<0.05)),frontal area ASM(r=0.224,P<0.05),left cheek ASM(r=0.333,P<0.01),nasal area ASM(r=0.239,P<0.05),mandibular area ASM(r=0.238,P<0.05),frontal area mean(r=0.251,P<0.05),nasal area mean(r=0.190,P<0.05),mandibular region mean(r=0.193,P<0.05),frontal region median(r=0.269,P<0.05),left cheek median(r=0.216,P<0.05),and right cheek median(r=0.203,P<0.05)were positively correlated with spleen,and frontal region a-value(r=-0.190,P<0.05),left cheek a-value(r=-0.187,P<0.05),mandibular region a-value(r=-0.203,P<0.05),left cheek CON(r=-0.275,P<0.01),nasal region CON(r=-0.265,P<0.01),mandibular region CON(r=-0.237,P<0.05),left cheek ENT(r=-0.245,P<0.01)(r=-0.247,P<0.01),left cheek mean(r=-0.191,P<0.05),and frontal area SD(r=-0.195,P<0.05)were negatively correlated with the spleen;three parameters had correlation with the hepatic evidence score,the right cheek L value(r=0.191,P<0.05)was positively correlated with the liver,and the right cheek a value(r=-0.188,P<0.05)and nasal a-value(r=-0.182,P<0.05)were negatively correlated with liver;8 parameters were correlated with renin score,mandibular a-value(r=0.527,P<0.05),left cheek b-value(r=0.458,P<0.05)and right cheek b-value(r=0.456,P<0.05)were positively correlated with kidney.The mandibular region L value(r=-0.500,P<0.05),nasal region COR(r=-0.501,P<0.05),right cheek mean(r=-0.443,P<0.05),mandibular region mean(r=-0.484,P<0.05),and mandibular region intermediate(r=-0.471,P<0.05)were negatively correlated with the kidney.(2)Parameters associated with the pathological evidence elements: 37 parameters were obtained in the pathological evidence elements.A total of 21 parameters were obtained related to the integral of qi stagnation evidence elements,including left cheek L value(r=0.207,P<0.01),right cheek L value(r=0.236,P<0.01),mandibular region b value(r=0.158,P<0.05),frontal region ASM(r=0.191,P<0.05),left cheek ASM(r=0.269,P<0.01),right cheek ASM(r=0.254,P<0.01),nasal ASM(r=0.269,P<0.01),mandibular ASM(r=0.299,P<0.01),frontal COR(r=0.181,P<0.05),mandibular COR(r=0.202,P<0.01),left cheek mean(r=0.164,P<0.05),right cheek mean(r=0.193,P<0.05),left cheek median(r=0.176,P<0.05),right cheek median(r=0.209,P<0.01)were positively correlated with air stagnation,left cheek a-value(r=-0.167,P<0.05),right cheek a-value(r=-0.153,P<0.05),left cheek CON(r=-0.183,P<0.05),mandibular area CON(r=-0.296,P<0.01),left cheek ENT(r=-0.205,P<0.01),right cheek ENT(r=-0.148,P<0.05),and mandibular area ENT(r=-0.202,P<0.05)were negatively correlated with qi stagnation;there were seven parameters related to the wet evidence element integral.Among them,left cheek L value(r=0.206,P<0.01),nasal area L value(r=0.198,P<0.01),mandibular area L value(r=0.187,P<0.05),frontal area ASM(r=0.299,P<0.05),right cheek ASM(r=0.186,P<0.05),right cheek middle value(r=0.160,P<0.05)were positively correlated with wetness,and left cheek a-value(r=-0.152,P<0.05)was negatively correlated with wetness;there were four parameters correlated with Yang deficiency evidence element score,among which right cheek L-value(r=0.197,P<0.05)and right cheek middle value(r=0.184,P<0.05)were positively correlated with Yang deficiency,and nasal area CON(r=-0.215,P<0.05)and nasal area ENT(r=-0.215,P<0.05)were negatively correlated with Yang deficiency;there were five parameters related to the heat evidence element score,of which the mean value of the mandibular region(r=0.228,P<0.05)and the middle value of the mandibular region(r=0.222,P<0.05)were positively correlated with heat,and the L value of the nasal region(r=-0.299,P<0.01)and the mean value of the nasal region(r=-0.275,P<0.01),and the middle value of the nasal region(r=-0.289,P<0.01)were negatively correlated with heat;six parameters were correlated with the phlegm evidence score,including the left cheek ASM(r=0.258,P<0.05),mandibular region ASM(r=0.266,P<0.01),frontal region COR(r=0.263,P<0.05),left cheek COR(r=0.214,P<0.05),ENT in the nasal region(r=0.299,P<0.05)were positively correlated with phlegm,and left cheek CON(r=-0.312,P<0.01)was negatively correlated with phlegm;no parameters were found to be correlated with the Yin deficiency evidence element score;a total of 11 parameters were correlated with the Qi deficiency evidence element score,of which L value in the frontal region(r=0.270,P<0.05),left cheek L value(r=0.270,P<0.05),right cheek L value(r=0.349,P<0.01),nasal ASM(r=0.272,P<0.05),frontal mean(r=0.236,P<0.05),right cheek mean(r=0.319,P<0.01),middle frontal value(r=0.267,P<0.05),right cheek middle value(r=0.312,P<0.05)were positively correlated with qi deficiency,left cheek a-value(r=-0.363,P<0.01),right cheek a-value(r=-0.337,P<0.01),nasal area a-value(r=-0.288,P<0.05)were negatively correlated with qi deficiency;one parameter was correlated with food accumulation evidence element score,frontal area ENT(r=0.419,P<0.05)was positively correlated with food stasis;1 parameter was correlated with blood stasis;SD value of mandibular region(r=-0.417,P<0.05)was negatively correlated with blood stasis.Conclusions1.Patients with chronic gastritis are mainly located in the stomach and are closely related to the liver and spleen.In terms of pathology,both CNAG and CAG are characterized by a mixture of deficiency and actuality,with qi stagnation,dampness and yang deficiency being important pathological features of CNAG and dampness,qi stagnation and phlegm condensation being important pathological features of CAG.2.The facial color and shape parameters of patients with chronic gastritis are characterized by dark red,dark yellow,and less luster,finer,deeper,and more dense skin lines,more scattered skin direction,and more number of wrinkles compared to healthy people.Facial diagnosis in TCM can provide reference for early screening of chronic gastritis and diagnosis of the disease.3.The color and shape parameters of different facial regions have a certain correlation with the location and disease of patients with chronic gastritis,among which the parameters related to spleen and qi stagnation are the most,and the facial color and shape parameters can reflect the pathological characteristics of the disease. |
PDF Full Text Request