3139

View Abstract / View Presentation

Locating seed automatically in posterior cingulate cortex for resting state fMRI data analysis by using unsupervised machine learning

Mingyi Li¹, Katherine Koenig¹, Jian Lin¹, and Mark Lowe¹
¹Imaging Institute, Cleveland Clinic, Cleveland, OH, United States

We developed a fully automatic data-driven method to generate seed clusters and corresponding maps for rs-fMRI data analysis by using machine learning. This method could generate seed in PCC to match the manually picked seed as long as the seed searching ROI included the manually picked seed.

Figure 2. Matched seeds. Top row shows manually picked seed and bottom row shows automatically generated seed.

Figure 1. Generating feature vector by combining rs-fMRI connectivity and T1 parcellation. Panel A: Z-map, Panel B: Z-score distribution, Panel C: Freesurfer parcellation, Panel D: feature vectors.