Estimating subtle cord atrophy is important in MS.
Recent work [LINK TO PAUL REPOS] introduced a pipeline to quantify the sensitivity of "hard" segmentation to detect small atrophies. The pipeline is based on a global resizing of the image.
With the recent introduction of SoftSeg [REF], which produces soft segmentations, we would like to investigate whether CSA measured from soft segmentation is more sensitive to detect subtle atrophy.
Context
Estimating subtle cord atrophy is important in MS. Recent work [LINK TO PAUL REPOS] introduced a pipeline to quantify the sensitivity of "hard" segmentation to detect small atrophies. The pipeline is based on a global resizing of the image. With the recent introduction of SoftSeg [REF], which produces soft segmentations, we would like to investigate whether CSA measured from soft segmentation is more sensitive to detect subtle atrophy.
Methodology
Bibliography