Diagnosis of MSA-C, MSA-P, and IPD by heterogeneity of brain metabolism in [18F] FP-CIT images

37 Objectives The quantification of heterogeneity for the striatum with 18F FP-CIT PET images gives a clinical prediction guide for distinguishing between patients with multiple system atrophy with predominant parkinsonism (MSA-P), MSA with cerebellar features (MSA-C), and idiopathic Parkinson’s disease (IPD). Methods The volume of the striatal region was defined according to an equi-contour line around SUV 2.5 g/ml for each region of striatum by PMod 3.6. The values of global and textural feature parameters (local and regional algorithms) were analyzed within VOI of 90 patients with non-PD, MSA-P, MSA-C, and IPD. The volume of the striatal region was defined according to an equi-contour line around SUV 2.5 g/ml for each region of striatum by PMod 3.6. The global parameters of SUVmax, SUVmean, SUVSD, SUV skewness, SUV kurtosis, TLG, tumor volume, entropy, SUV peak were calculated. Fifty-nine parameters from textural feature analysis were calculated using a co‐occurrence matrix, a voxel‐alignment matrix, a neighborhood intensity difference matrix, an intensity size‐zone matrix, a normalized co‐occurrence matrix, a texture spectrum, texture feature coding, a texture feature coding co‐occurrence matrix and neighborhood gray‐level dependence. Results Seventy-two parameters related to heterogeneity were analyzed to distinguish between patients with MSA-P, MSA-C, and IPD, by using global and textural feature analysis. Most of the parameters (more than 54 -- especially size-zone variability, code similarity , contrast, inverse difference moment and homogeneity) could distinguish between normal patients and those with IPD (or MSA-P) by the heterogeneity when the p-value is less than 0.05, but could not distinguish between normal patients and those with MSA-C.The 2 parameters (high intensity zone emphasis and low intensity short-zone emphasis) of heterogeneity were distinguished among patients with normal, IDP, and MSA-P with p Conclusions Heterogeneity of 18F FP-CIT PET in the striatal region images is a prognostic factor for distinguishing between patients with normal, IDP, and MSA-P. Fig 1. Statistical comparison of the heterogeneities by two sample t-test between idiopathic Parkinson’s disease (IPD) and non-PD are presented by box plot. The p-values are calculated by R-statistics package.