PURPOSE: To investigate the diffusion time dependency of water diffusion in cortical brain tissue. METHODS: We have combined an oscillating gradient spin-echo (OGSE) and a pulse gradient spin echo (PGSE) spin-echo sequence to acquire diffusion-weighted MRI images in vivo in healthy rat brains over a wide range of diffusion times (1.9-29.2 ms) and estimated the parameters of the biexponential and cumulant expansion diffusion MRI signal models. Diffusion images were obtained at 17.2 Tesla with maximum gradient strength of 1000 mT/m allowing 40 b values up to approximately 4000 s/mm(2). RESULTS: At all diffusion times the log plot of diffusion signal attenuation versus b value was curved, confirming that diffusion is not free, even at very short diffusion times. This suggests that the length scale of obstacles to diffusion must be smaller than the corresponding shortest observed diffusion distance (approximately 1.7 μm). The diffusion MRI signal was also not found in a steady-state, even at our longest diffusion time (29.2 ms), suggesting some degree of segregation of water in pools. CONCLUSION: Overall, the results showed that the parameters derived from the two diffusion models could not well be related to specific tissue features. More specific models must be developed taking into account diffusion signal behavior at high b values and short diffusion times.
PURPOSE: To investigate the diffusion time dependency of water diffusion in cortical brain tissue. METHODS: We have combined an oscillating gradient spin-echo (OGSE) and a pulse gradient spin echo (PGSE) spin-echo sequence to acquire diffusion-weighted MRI images in vivo in healthy rat brains over a wide range of diffusion times (1.9-29.2 ms) and estimated the parameters of the biexponential and cumulant expansion diffusion MRI signal models. Diffusion images were obtained at 17.2 Tesla with maximum gradient strength of 1000 mT/m allowing 40 b values up to approximately 4000 s/mm(2). RESULTS: At all diffusion times the log plot of diffusion signal attenuation versus b value was curved, confirming that diffusion is not free, even at very short diffusion times. This suggests that the length scale of obstacles to diffusion must be smaller than the corresponding shortest observed diffusion distance (approximately 1.7 μm). The diffusion MRI signal was also not found in a steady-state, even at our longest diffusion time (29.2 ms), suggesting some degree of segregation of water in pools. CONCLUSION: Overall, the results showed that the parameters derived from the two diffusion models could not well be related to specific tissue features. More specific models must be developed taking into account diffusion signal behavior at high b values and short diffusion times.
Authors: F Natali; C Dolce; J Peters; C Stelletta; B Demé; J Ollivier; M Boehm; G Leduc; I Piazza; A Cupane; E L Barbier Journal: J R Soc Interface Date: 2019-08-14 Impact factor: 4.118