The segmentation of human hippocampal subfields on in vivo MRI has gained great interest in the last decade, because these anatomic subregions were found to be highly specialized in recent studies and are potentially affected differentially by normal aging, Alzheimer’s disease, schizophrenia, epilepsy, major depressive disorder, and posttraumatic stress disorder. However, manually segmenting hippocampal subfields is labor-intensive and time-consuming, which limits the study to a small sample size. We developed a multi-scale Convolutional neural network based Automated hippocampal subfield Segmentation Toolbox (CAST) for automated segmentation, which can be easily trained and output segmented images in one minute.
This abstract and the presentation materials are available to members only; a login is required.