The efficiency of retrospective artifact correction methods in improving the statistical power of between-group differences in spinal cord DTI

Diffusion tensor imaging (DTI) is a technique used to examine the white matter microstructure of the spinal cord, but it's prone to various artifacts. Retrospective correction techniques are used to reduce these artifacts without increasing scan time. This paper introduces a novel outlier rejection method called 'reliability masking' to improve existing correction approaches. It excludes unreliable data points from DTI index maps. The study investigates how different combinations of retrospective correction techniques, including registration, robust fitting, and reliability masking, affect the statistical power of findings related to cervical spondylotic myelopathy (CSM) patients.

• 1
  Adding reliability masking to established post-processing steps increased the statistical power of clinical findings by 4.7%.
• 2
  Reliability masking primarily improved statistical power by reducing variability within the study groups.
• 3
  Registration, while used for correction, slightly increased variability, contrasting with reliability masking's effect.