The purpose of this project is to develop fully automatic pipeline for fetal brain MRI processing including automatic brain segmentation; fetal MRI quality assessment; motion correction; fetal brain tissue segmentation (Hong et al., Frontiers in Neuroscience, 2020); cortical surface extraction; and volume- and surface-based feature extraction. We are using state-of-the-art medical image processing and deep learning techniques for this pipeline development.

We developed a novel method for automatic labeling of cortical sulci on the fetal cortical surface. We use the weighted sulcal probability maps from the multiple templates and adopt sulcal basin-wise approach to assign sulcal labels to each basin. The mean accuracy of our approach was 0.958 across subjects, significantly higher than the accuracies of the other approaches. This novel approach shows highly accurate sulcal labeling and provides a reliable means to examine characteristics of cortical regions in the fetal brain (Yun et al., Neuroimage 2019).

For more local cortical analysis at a vertex level, we define vertex spatial correspondence using 2D surface registration tool across different fetal brains with different gestational age.