• Yasuhiko Tachibana, Takayuki Obata, Mariko Yoshida, Masaaki Hori, Koji Kamagata, Michimasa Suzuki, Issei Fukunaga, Kouhei Kamiya, Kazumasa Yokoyama, Nobutaka Hattori, Tomio Inoue, and Shigeki Aoki.
• Research Center for Charged Particle Therapy, National Institute of Radiological Sciences, 4-9-1 Anagawa, Inage-ku, Chiba, 263-8555, Japan.
• Eur Radiol. 2015 Jun 1; 25 (6): 1701-7.

ObjectivesTo compare the significance of the two-compartment model, considering diffusional anisotropy with conventional diffusion analyzing methods regarding the detection of occult changes in normal-appearing white matter (NAWM) of multiple sclerosis (MS).MethodsDiffusion-weighted images (nine b-values with six directions) were acquired from 12 healthy female volunteers (22-52 years old, median 33 years) and 13 female MS patients (24-48 years old, median 37 years). Diffusion parameters based on the two-compartment model of water diffusion considering diffusional anisotropy was calculated by a proposed method. Other parameters including diffusion tensor imaging and conventional apparent diffusion coefficient (ADC) were also obtained. They were compared statistically between the control and MS groups.ResultsDiffusion of the slow diffusion compartment in the radial direction of neuron fibers was elevated in MS patients (0.121 × 10(-3) mm2/s) in comparison to control (0.100 × 10(-3) mm(2)/s), the difference being significant (P = 0.001). The difference between the groups was not significant in other comparisons, including conventional ADC and fractional anisotropy (FA) of diffusion tensor imaging.ConclusionThe proposed method was applicable to clinically acceptable small data. The parameters obtained by this method improved the detectability of occult changes in NAWM compared to the conventional methods.Key Points• Water diffusion was compared between the controls and multiple sclerosis patients. • A two-compartment model, considering diffusional anisotropy was selected for water diffusion analysis. • Axial and radial diffusion of fast and slow diffusion components were evaluated. • A new method was developed to obtain the metrics stably. • The metrics indicated high detectability of slight differences between the groups.

19 keywords...

hide…