2D distribution analysis of DTI in two phenotypes of dystonia patients

Introduction Hereditary dystonia is a neurological movement disorder where the subjects have abnormal motions due to muscle contractions [1]. Only a subset of the patients who carry the gene show the overt symptoms [2]. Although the pathology is not completely understood, purkinje cells, basal ganglia and part of the motor circuit are all implicated in this disorder [3, 4]. According to the diffusional characteristics, diffusion tensor imaging can be used to characterize normal and pathological tissues [5, 6]. We propose to use a 2D distribution analysis with a physical brain model, which can automatically determine the different tissue types according the inherent diffusional characteristics [7]. Development of this type of analysis may be used to help to diagnose non-focal disease state. Subjects and Methods Three groups of subjects were studied: (1) Seven dystonia patients with DYT1 genotype (4M/3F, age 39.8±19.4); (2) four non-manifesting DYT1 mutation carriers (2M/2F, age 46.9±15.4); and (3) eight normal subjects (5M/3F, age 50.1±10.6) [4]. Images were acquired in a clinical 3T scanner. A single-shot spinecho EPI sequence was used with 55 isotropically distributed diffusion gradient directions and one b0 image. The b-value in the DWIs was 1000 s/mm. The DTI protocol included 72 slices with 1.8mm thickness, FOV 23cm, TE 68.3 ms and TR 7s. Images were zero filled to a matrix size of 256 × 256, yielding an image resolution of 0.9 × 0.9× 1.8 mm. After data acquisition, Dav (=Trace/3), FA maps are calculated. Using a physical diffusion brain model of four compartments, 2-D distribution map was fitted using four 2-D Gaussian functions as: