Multi-atlas segmentation is an effective approach for automatically labeling objects of interest in biomedical images. In this approach, multiple expert-segmented example images, called atlases, are registered to a target image, and deformed atlas segmentations are combined using label fusion. Among the proposed label fusion strategies, weighted voting with spatially varying weight distributions derived from atlas-target intensity similarity have been particularly successful. However, one limitation of these strategies is that the weights are computed independently for each atlas, without taking into account the fact that different atlases may produce similar label errors. To address this limitation, we propose a new solution for the label fusion problem, in which weighted voting is formulated in terms of minimizing the total expectation of labeling error, and in which pairwise dependency between atlases is explicitly modeled as the joint probability of two atlases making a segmentation error at a voxel. This probability is approximated using intensity similarity between a pair of atlases and the target image in the neighborhood of each voxel. We validate our method in two medical image segmentation problems: hippocampus segmentation and hippocampus subfield segmentation in magnetic resonance (MR) images. For both problems, we show consistent and significant improvement over label fusion strategies that assign atlas weights independently.

We propose a simple but generally applicable approach to improving the accuracy of automatic image segmentation algorithms relative to manual segmentations. The approach is based on the hypothesis that a large fraction of the errors produced by automatic segmentation are systematic, i.e., occur consistently from subject to subject, and serves as a wrapper method around a given host segmentation method. The wrapper method attempts to learn the intensity, spatial and contextual patterns associated with systematic segmentation errors produced by the host method on training data for which manual segmentations are available. The method then attempts to correct such errors in segmentations produced by the host method on new images. One practical use of the proposed wrapper method is to adapt existing segmentation tools, without explicit modification, to imaging data and segmentation protocols that are different from those on which the tools were trained and tuned. An open-source implementation of the proposed wrapper method is provided, and can be applied to a wide range of image segmentation problems. The wrapper method is evaluated with four host brain MRI segmentation methods: hippocampus segmentation using FreeSurfer (Fischl et al., 2002); hippocampus segmentation using multi-atlas label fusion (Artaechevarria et al., 2009); brain extraction using BET (Smith, 2002); and brain tissue segmentation using FAST (Zhang et al., 2001). The wrapper method generates 72%, 14%, 29% and 21% fewer erroneously segmented voxels than the respective host segmentation methods. In the hippocampus segmentation experiment with multi-atlas label fusion as the host method, the average Dice overlap between reference segmentations and segmentations produced by the wrapper method is 0.908 for normal controls and 0.893 for patients with mild cognitive impairment. Average Dice overlaps of 0.964, 0.905 and 0.951 are obtained for brain extraction, white matter segmentation and gray matter segmentation, respectively.

Generalized ligamentous laxity is an independent predictor of poor outcomes and a risk factor of recurrent instability following the modified Broström procedure for CLAI.

Automatic segmentation using multi-atlas label fusion has been widely applied in medical image analysis. To simplify the label fusion problem, most methods implicitly make a strong assumption that the segmentation errors produced by different atlases are uncorrelated. We show that violating this assumption significantly reduces the efficiency of multi-atlas segmentation. To address this problem, we propose a regression-based approach for label fusion. Our experiments on segmenting the hippocampus in magnetic resonance images (MRI) show significant improvement over previous label fusion techniques.

The hippocampal formation (HF) is a brain structure of great interest because of its central role in learning and memory, and its associated vulnerability to several neurological disorders. In vivo oblique coronal T2-weighted MRI with high in-plane resolution (~0.5 mm×0.5 mm), thick slices (~2.0 mm), and a field of view tailored to imaging the hippocampal formation (denoted HF-MRI in this paper) has been advanced as a useful imaging modality for detailed hippocampal morphometry. Cross-sectional analysis of volume measurements derived from HF-MRI has shown the modality’s promise to yield sensitive imaging-based biomarker for neurological disorders such as Alzheimer’s disease. However, the utility of this modality for making measurements of longitudinal change has not yet been demonstrated. In this paper, using an unbiased deformation-based morphometry (DBM) pipeline, we examine the suitability of HF-MRI for estimating longitudinal change by comparing atrophy rates measured in the whole hippocampus from this modality with those measured from more common isotropic (~1 mm3) T1-weighted MRI in the same set of individuals, in a cohort of healthy controls and patients with cognitive impairment. While measurements obtained from HF-MRI were largely consistent with those obtained from T1-MRI, HF-MRI yielded slightly larger group effect of greater atrophy rates in patients than in controls. The estimated minimum sample size required for detecting a 25% change in patients’ atrophy rate in the hippocampus compared to the control group with a statistical power β=0.8 was N=269. For T1-MRI, the equivalent sample size was N=325. Using a dataset of test–retest scans, we show that the measurements were free of additive bias. We also demonstrate that these results were not a confound of certain methodological choices made in the DBM pipeline to address the challenges of making longitudinal measurements from HF-MRI, using a region of interest (ROI) around the HF to globally align serial images, followed by slice-by-slice deformable registration to measure local volume change. Additionally, we present a preliminary study of atrophy rate measurements within hippocampal subfields using HF-MRI. Cross-sectional differences in atrophy rates were detected in several subfields.