Font Size: a A A

Correlation Analysis Between Microvessel Density And CT Features Of Ground Glass Nodule By Using Image-Pro Plus Combined With Photoshop Image Analysis Software

Posted on:2019-05-08Degree:MasterType:Thesis
Country:ChinaCandidate:J ZhaoFull Text:PDF
GTID:2394330545497542Subject:Medical imaging and nuclear medicine
Abstract/Summary:PDF Full Text Request
Objective: To explore the value of Image-Pro Plus 6.0(IPP 6.0)combined with Photoshop image analysis software to quantitatively measure the microvessel density(MVD)of ground glass nodule(GGN),and analyze the relationship between CT features and MVD value,MVD stage of GGN,to better understanding the pathological basis of GGN and provide a theoretical basis for clinical diagnosis,treatment and prognostic estimation.Materials and Methods: Seventy-seven cases of lung adenocarcinomas were recruited from Affiliated Zhongshan Hospital of Dalian University during January 2014 to December 2017,including 23 pre-invasive cases,22 minimally invasive adenocarcinomas(MIA)cases and 32 invasive adenocarcinomas(IA)cases.All cases accepted surgery and were pathologically proved as adenocarcinoma.They were divided into non-invasive lesion group(AAH 2 cases,AIS 21 cases)and invasive lesion group(MIA 22 cases,IA 32 cases),according to whether there were infiltrative lesions in GGN pathology when the expression of MVD in the two groups were evaluated and the correlation between CT features and the expression of MVD in the two groups were analyzed.The diameter of all the cases was less than 3cm,ground glass opacity component was more than 50%.And all patients did not accept biopsy or radio-chemotherapy before preoperative CT examination.A routine hematoxylin and eosin staining,and immune-histochemical stain with a mouse monoclonal antibody CD31 were performed to all the cases.The counting method is as follows: each slide was first scanned at low magnification(×100)to identify three “hot spot” areas where the most microvessels was shown the best.Subsequently,these areas were photographed separately at high magnification(×200).And these pictures were analyzed by using IPP 6.0 combined with Photoshop image analysis software to automatically measure the value of MVD of GGN.Any brown-staining endothelial cell or clusters that were clearly separated from one another and no matter whether there was a lumen or not were considered as a single microvessel.The microvessel of lumen diameter with more than eight red blood cells,blood vessels with thicker muscular layer and some microvessel in benign lesion areas,less tumor cells areas and tumor necrosis area should be exclude.We defined a degree corresponding to the value of the MVD: Degree ?,mild vascular proliferation(MVD?15),Degree ?,moderate vascular proliferation(15<MVD<25),Degree ?,obvious vascular proliferation(MVD?25).The CT features including lesion location,the type of density,lobulation sign,speculation,vascular ingress,pleural indentation,bubble-like lucency,air bronchogram and mean diameter,mean CT attenuation were analyzed retrospectively.Statistical analysis was performed using SPSS 22.0 software.One-way analysis of variance(ANOVA)was used to compare differences of MVD value of GGN among pre-invasive lesions,MIA and IA groups,and Chi-square test was used to compare MVD stage between three groups.The independent t-test were used to compare differences of MVD value,mean diameter,mean CT attenuation of GGN among non-invasive lesion group and invasive lesion group,and Chi-square test or Fisher exact test was used to compare CT morphology features,MVD stage between two groups.Pearson correlation analysis was used to evaluate the correlation between lesion size,mean CT attenuation and MVD value,and the corresponding scatter plot were also drawn.Spearman correlation analysis was used to analyze the correlation between MVD stage and all CT feature indexes.A P-value less than 0.05 was considered as statistically significant.Results:1.There were statistically significant differences in MVD value among pre-invasive lesions,MIA and IA groups.And the MVD value displayed gradually increasing from pre-invasive lesions to MIA,IA group,27.70±10.59 number per field,36.91±12.73 number per field,47.01±15.02 number per field respectively.There were also statistically significant differences in the MVD stage of the lesions among three groups(P<0.001),about 96.9% IA exist more obvious vascular proliferation(degree?),which was much higher than pre-invasive lesions group(47.8%)and MIA group(63.7%).2.The MVD value of invasive lesion group was significantly higher than that of non-invasive lesion group,42.89±14.87 number per field,27.70±10.59 number per field(P<0.001)respectively.In terms of MVD stage,invasive lesion group(83.3%)exist more obvious vascular proliferation(degree?)than non-invasive lesion group(47.8%),and the difference was statistically significant(P<0.05).3.There were statistically significant differences in the type of density,lobulation,speculation,air bronchogram,mean diameter and mean CT attenuation between the non-invasive lesion group and invasive lesion group group(P<0.05).The type of density,lobulation sign of GGN was closely correlated with MVD value(r=0.414,P=0.002;r=0.373,P=0.005).However,the MVD stage-related CT feature was only lubulation sign(r=0.330,P=0.015).The mean diameter of GGN had moderate positive correlated with MVD values,MVD stage(r=0.494,P<0.001;r=0.305,P=0.025).The mean CT attenuation of GGN had also moderate positive correlated with MVD values(r=0.300,P=0.028),but it failed to yield a significant association with MVD stage(r=0.127,P=0.359).Conclusion:1.IPP 6.0 combined with Photoshop image analysis software is a new accurate method with higher efficiency to measure the expression of MVD of GGN,not only it is easy to handle,but also the data can be automatically counted and generated,which could be a new method to quickly and accurately measure the content of MVD(MVD value)and MVD stage of GGN,and is worth promoting the clinical application.2.Some CT features of GGN were strong correlated with the expression of MVD,which might be an indicator to identify the pathology types,evaluate the invasion of lung adenocarcinoma with GGN and provide the power information,imaging basis for clinical diagnosis,treatment and prognostic evaluation.
Keywords/Search Tags:Image-Pro Plus, Photoshop image analysis software, ground glass nodule, microvessel density, computed tomography
PDF Full Text Request
Related items