Can we use k-d Tree Based k-Means Clustering in a 3D vector?

I am segmenting the spinal canal in CT images and I need to classify the spinal canal pixels positions in groups.
( “Automatic Segmentation of the Body and the Spinal Canal in CT Images Based on A Priori Information”).
Can I use ITK’s k-d Tree Based k-Means Clustering ??

Thanks in advance,

Rafael Scatena