ANIMAL+INSECT: Improved Cortical Structure Segmentation

An algorithm for improved automatic segmentation of gross anatomical structures of the human brain is presented that merges the output of a tissue classification process with gross anatomical region masks, automatically defined by non-linear registration of a given data set with a probabilistic anatomical atlas. Experiments with 20 real MRI volumes demonstrate that the method is reliable, robust and accurate. Manually and automatically defined labels of specific gyri of the frontal lobe are similar, with a Kappa index of 0.657.

• Type: Conference
• Year: 1999
• Where: IPMI
• Authors: D. Louis Collins, Alex P. Zijdenbos, Wim F. C. Baaré, Alan C. Evans
• Keywords: Gross Anatomical Region, Gross Anatomical Structures, IPMI 1999, Medical Imaging, Probabilistic Anatomical Atlas