Kyle Taljan1,2, Cameron C. McIntyre1,
Ken E. Sakaie3
1Lerner Research Institute,
Biomedical Engineering, Cleveland Clinic, Cleveland, OH, United States; 2Biomedical
Engineering, Cleveland State University, Cleveland, OH, United States; 3Imaging
Institute, Cleveland Clinic
Pioneering work by Behrens et al demonstrated the potential of high angular resolution diffusion imaging (HARDI) for segmenting brain structures based on their anatomical connectivity. The methodology uses a winner-takes-all approach in which classification does not distinguish between slight or large differences in connectivity. However, secondary connections may be important but are masked by the winner-takes-all approach. We demonstrate this point by examining connectivity of the internal capsule (IC), a potentially significant target for deep brain stimulation (DBS) treatment of refractory depression.