Comparative Evaluation Of Three Methods Of Automatic Segmentation Of\n Brain Structures Using 426 Cases

Segmentation of brain structures in a large dataset of magnetic resonance\nimages (MRI) necessitates automatic segmentation instead of manual tracing.\nAutomatic segmentation methods provide a much-needed alternative to manual\nsegmentation which is both labor intensive and time-consuming. Among brain\nstructures, the hippocampus presents a challenging segmentation task due to its\nirregular shape, small size, and unclear edges. In this work, we use\nT1-weighted MRI of 426 subjects to validate the approach and compare three\nautomatic segmentation methods: FreeSurfer, LocalInfo, and ABSS. Four\nevaluation measures are used to assess agreement between automatic and manual\nsegmentation of the hippocampus. ABSS outperformed the others based on the Dice\ncoefficient, precision, Hausdorff distance, ASSD, RMS, similarity, sensitivity,\nand volume agreement. Moreover, comparison of the segmentation results,\nacquired using 1.5T and 3T MRI systems, showed that ABSS is more sensitive than\nthe others to the field inhomogeneity of 3T MRI.\n