DTI-based fiber tractography: why not?

Abstract

Diffusion tensor imaging (DTI) has been fundamental in enabling the noninvasive exploration of the brain white matter architecture. However, the use of DTI for tractography remains challenging due to the known limitations of the tensor model in capturing complex fiber configurations. This short communication summarizes the key points of a debate held at the 2024 Tract-Anat Retreat sparked by the provocative statement: “it is never acceptable to use DTI for tractography”. While we identified the advantages of moving towards more advanced orientation models that can provide more complete and accurate reconstructions of white matter pathways, we also highlighted the valuable contribution DTI-based tractography can make in clinical contexts, or when investigating simpler fiber architectures. These perspectives underscore the importance of the specific application when evaluating whether it is acceptable or not to use the tensor model for tractography.