Performing label-fusion-based segmentation using multiple automatically generated templates

TLDR

Classical model‑based MRI segmentation aligns images to a single atlas, but accuracy suffers from atlas bias, misregistration, and resampling error, while multi‑atlas methods mitigate this by matching to several manually labeled templates and fusing results through voxel‑wise voting. The study aims to extend multi‑atlas segmentation to use automatically generated template libraries (MAGeT Brain) for unique, time‑consuming atlases. The method applies MAGeT Brain to mouse and human data, employing ANIMAL and ANTs registrations on a high‑resolution mouse atlas and a histology‑derived human basal ganglia/thalamus atlas. MAGeT Brain achieved Dice κ = 0.801 for the mouse anterior commissure (with a possible ceiling for hippocampus) and improved human subcortical segmentation (κ = 0.845, 0.752, 0.861 for striatum, globus pallidus, thalamus), matching or surpassing multi‑atlas accuracy (κ = 0.894, 0.815, 0.895).

Abstract

Classically, model-based segmentation procedures match magnetic resonance imaging (MRI) volumes to an expertly labeled atlas using nonlinear registration. The accuracy of these techniques are limited due to atlas biases, misregistration, and resampling error. Multi-atlas-based approaches are used as a remedy and involve matching each subject to a number of manually labeled templates. This approach yields numerous independent segmentations that are fused using a voxel-by-voxel label-voting procedure. In this article, we demonstrate how the multi-atlas approach can be extended to work with input atlases that are unique and extremely time consuming to construct by generating a library of multiple automatically generated templates of different brains (MAGeT Brain). We demonstrate the efficacy of our method for the mouse and human using two different nonlinear registration algorithms (ANIMAL and ANTs). The input atlases consist a high-resolution mouse brain atlas and an atlas of the human basal ganglia and thalamus derived from serial histological data. MAGeT Brain segmentation improves the identification of the mouse anterior commissure (mean Dice Kappa values (κ = 0.801), but may be encountering a ceiling effect for hippocampal segmentations. Applying MAGeT Brain to human subcortical structures improves segmentation accuracy for all structures compared to regular model-based techniques (κ = 0.845, 0.752, and 0.861 for the striatum, globus pallidus, and thalamus, respectively). Experiments performed with three manually derived input templates suggest that MAGeT Brain can approach or exceed the accuracy of multi-atlas label-fusion segmentation (κ = 0.894, 0.815, and 0.895 for the striatum, globus pallidus, and thalamus, respectively).