A Study Of Risk Factors For White Matter Hyperintensity In The Aged And The Application Of Radiomics In Predicting The Progress Of White Matter Hyperintensity

Part Ⅰ:A Study of risk factors for White Matter Hyperintensities in old adults:Relationship with Hyperintensities Vessels Sign on MR FLAIR imagingObjective To investigate the correlation between white matter hyperintensities(WMH)in old adults and hyperintensities vessels sign(HVS)on MR FLAIR imaging,to explore the risk factors and pathogeneses of WMH.Methods Retrospectively collected image and clinical data of patients who performed both head-and-neck CTA and brain MRI examination within one month in our hospital during 2013-2016.Fazekas visual scale was used to evaluate periventricular white matter hyperintensity(PWMH)and deep white matter hyperintensity(DWMH)in both hemispheres.According to the presence or absence of HVS in cerebral hemisphere,it is divided into HVS positive and negative groups.Compare clinical data,PWMH,and DWMH differences between two groups.Results A total of 271 patients,542 cerebral hemispheres,were included in the study.HVS was positive in 79(14.6%)and negative in 463(85.4%)hemispheres.There was a significant difference between HVS positive and negative groups in the ipsilateral CIA stenosis(X~2=126.840,P<0.01).The incidence of ipsilateral severe carotid artery stenosis in HVS positive group was 62.0%(49/79),significantly higher than 9.9%(46/463)in the HVS negative side.The occurrence of moderate-severe DWMH was65.8%(52/79)in HVS positive group,while 34.8%(161/463)in negative group.There was a significant difference between the two groups(X~2=34.962,P<0.01).But the occurrence of moderate-severe PWMH in the two groups was 65.8%(52/79)and55.5%(257/463),respectively,without significant difference(X~2=6.944,P=0.074).After adjusting for age,gender,ipsilateral ICA stenosis,hypertension,Diabetes,etc.,multivariate analysis revealed that HVS positive still was the independent risk factor of DWMH(OR=2.653,95%CI 1.489-4.726,P=0.001).Conclusion HVS positive was the independent risk factor of DWMH in the elderly,but there was no clear correlation with PWMH.It suggests that hypoperfusion is a possible pathogenesis of DWMH in the elderly.Part Ⅱ: Predicting the development of normal-appearing white matter with radiomics in the aging brain: a longitudinal clinical studyBackground: Normal-appearing white matter(NAWM)refers to the normal,yet diseased matter around the white matter hyperintensities(WMH)on conventional MR images.Radiomics is an emerging quantitative imaging technique that provides additional information about the tissues when compared with traditional visual analysis.The aim of this study was to explore whether WMH could be predicted at the early stage of NAWM with a textural analysis in the general elderly population.Methods: Imaging data were retrieved from PACS from 2012 to 2017.The subjects(≥60y)received more than two MRI exams on the same machine with time intervals of more than one year.By comparing baseline and follow-up images,patients with noted progression of WMH were included as the case group(n=51),while age-matched subjects without WMH were included as the control group(n=51).Segmentations of the regions of interest(ROIs)were done with the ITK software.Two ROIs of developing NAWM(d NAWM)and non-developing NAWM(non-d NAWM)were drawn separately on the FLAIR images of each patient.d NAWM appeared normal on the baseline images,yet evolved into WMH on the follow-up images.Non-d NAWM appeared normal on both baseline and follow-up images.A third ROI of normal white matter(NWM)was extracted from normal control groups,which was normal on both baseline and follow-up images.Textural features were dimensionally reduced with ANOVA+MW,correlation analysis,and LASSO.Three models were built based on the optimal parameters of dimensional reduction,including Model 1(NWM vs.d NAWM),Model 2(non-d NAWM vs.d NAWM),and Model 3(NWM vs.non-d NAWM).The ROC curve was adopted to evaluate the classification validity of these models.Results: Basic characteristics of the patients and controls showed no statistically significant differences.The AUC of Model 1 in training and test groups were 0.967(95% CI: 0.831-0.999)and 0.954(95% CI: 0.876-0.989),respectively.The AUC of Model 2 were 0.939(95% CI: 0.856-0.982)and 0.846(95% CI: 0.671-0.950),respectively.The AUC of Model 3 were 0.713(95% CI: 0.593-0.814)and 0.667(95% CI: 0.475-0.825),respectively.Conclusions: Radiomics textural analysis can clearly distinguish d NAWM from non-d NAWM on FLAIR images,which could be used for the early detection of NAWM lesions before they develop into visible WHM.