In this manuscript, we investigate how application of established post-processing methods (spatial registration and robust tensor fitting) and of a newly introduced outlier rejection technique referred to as reliability masking influence the statistical power of a clinical spinal cord DTI study. The assessment was performed using a previously published clinical dataset investigating microstructural correlates of spinal degeneration in cervical spondylotic myelopathy (CSM). We found that the established post-processing methods had almost no influence on the statistical power by which the microstructural differences is observed, whereas reliability masking increased the statistical power by more than 13%.
This abstract and the presentation materials are available to members only; a login is required.