Belief function in low level data fusion: application in MRI images of vertebra

The work presented was done in order to help doctors to monitor patients with spinal diseases. The objective was the reconstruction of each vertebra of the lumbar spine from a series of parallel views. From an initial segmentation, we looked for the part of the image that best represents the vertebra´s anatomical contour, in order to give doctors a belief degree on each part of this segmentation. To find points on the cortex, low-level information is used, i.e. grey levels associated with spatial constraints. The originalities of this work are in the low-level fusion process, in the choice of the discriminating parameters for the expertise, and in the construction of the belief functions. This allows us to obtain the most reliable decisions, which are illustrated by experimental results.