Automatically Parcellating the Human Cerebral Cortex

TLDR

We present a technique that automatically assigns neuroanatomical labels to cortical surface locations using probabilistic information from a manually labeled training set. The method incorporates geometric data from the cortical model and neuroanatomical conventions from the training set, and is illustrated with examples from two different training sets, demonstrating its flexibility. The technique yields a complete labeling of cortical sulci and gyri and achieves accuracy comparable to manual labeling.

Abstract

We present a technique for automatically assigning a neuroanatomical label to each location on a cortical surface model based on probabilistic information estimated from a manually labeled training set. This procedure incorporates both geometric information derived from the cortical model, and neuroanatomical convention, as found in the training set. The result is a complete labeling of cortical sulci and gyri. Examples are given from two different training sets generated using different neuroanatomical conventions, illustrating the flexibility of the algorithm. The technique is shown to be comparable in accuracy to manual labeling.