Reducing computation time by Monte Carlo method: an application in determining axonal orientation distribution function

Abstract

Diffusion MRI (dMRI) is highly sensitive in detecting early cerebral ischemic changes in acute stroke, and in pre-clinical assessment of white matter (WM) anatomy using tractography, thus being an important component of health informatics. In clinical settings, the computation time is critical, and so finding forms of reducing the processing time in high computation processes such as Diffusion Spectrum Imaging (DSI) dMRI data processing is extremely relevant. We analyse here a method for reducing the computation of the dMRI-based axonal orientation distribution function h by using a Monte Carlo sampling-based methods for voxel selection, and so obtained a reduction in required data sampling of about 20%. In this work we show that the convergence to the correct value in this type of dMRI data-processing is linear and not exponential, implying that the Monte Carlo approach in this type of dMRI data processing improves its speed, but further improvements are needed.