Jennifer Alvén

Überatlas: Robust Speed-Up of Feature-Based Registration and Multi-Atlas Segmentation

Registration is a key component in multi-atlas approaches to medical image segmentation. Current state of the art uses intensity-based registration methods, but such methods tend to be slow, which sets practical limitations on the size of the atlas set. In this paper, a novel feature-based registration method for affine registration is presented. The algorithm constructs an abstract representation of the entire atlas set, an uberatlas, through clustering of features that are similar and detected consistently through the atlas set. This is done offline. At runtime only the feature clusters are matched to the target image, simultaneously yielding robust correspondences to all atlases in the atlas set from which the affine transformations can be estimated efficiently. The method is evaluated on 20 CT images of the heart and 30 MR images of the brain with corresponding gold standards. Our approach succeeds in producing better and more robust segmentation results compared to two baseline methods, one intensity-based and one feature-based, and significantly reduces the running times.

Überatlas: Fast and robust registration for multi-atlas segmentation☆

Multi-atlas segmentation has become a frequently used tool for medical image segmentation due to its outstanding performance. A computational bottleneck is that all atlas images need to be registered to a new target image. In this paper, we propose an intermediate representation of the whole atlas set – an uberatlas – that can be used to speed up the registration process. The representation consists of feature points that are similar and detected consistently throughout the atlas set. A novel feature-based registration method is presented which uses the uberatlas to simultaneously and robustly find correspondences and affine transformations to all atlas images. The method is evaluated on 20 CT images of the heart and 30 MR images of the brain with corresponding ground truth. Our approach succeeds in producing better and more robust segmentation results compared to three baseline methods, two intensity-based and one feature-based, and significantly reduces the running times.

Automatic pericardium segmentation and quantification of epicardial fat from computed tomography angiography

Abstract. Recent findings indicate a strong correlation between the risk of future heart disease and the volume of adipose tissue inside of the pericardium. So far, large-scale studies have been hindered by the fact that manual delineation of the pericardium is extremely time-consuming and that existing methods for automatic delineation lack accuracy. An efficient and fully automatic approach to pericardium segmentation and epicardial fat volume (EFV) estimation is presented, based on a variant of multi-atlas segmentation for spatial initialization and a random forest classifier for accurate pericardium detection. Experimental validation on a set of 30 manually delineated computer tomography angiography volumes shows a significant improvement on state-of-the-art in terms of EFV estimation [mean absolute EFV difference: 3.8 ml (4.7%), Pearson correlation: 0.99] with run times suitable for large-scale studies (52 s). Further, the results compare favorably with interobserver variability measured on 10 volumes.

Max-margin learning of deep structured models for semantic segmentation

During the last few years most work done on the task of image segmentation has been focused on deep learning and Convolutional Neural Networks (CNNs) in particular. CNNs are powerful for modeling complex connections between input and output data but lack the ability to directly model dependent output structures, for instance, enforcing properties such as smoothness and coherence. This drawback motivates the use of Conditional Random Fields (CRFs), widely applied as a post-processing step in semantic segmentation. In this paper, we propose a learning framework that jointly trains the parameters of a CNN paired with a CRF. For this, we develop theoretical tools making it possible to optimize a max-margin objective with back-propagation. The max-margin loss function gives the model good generalization capabilities. Thus, the method is especially suitable for applications where labelled data is limited, for example, medical applications. This generalization capability is reflected in our results where we are able to show good performance on two relatively small medical datasets. The method is also evaluated on a public benchmark (frequently used for semantic segmentation) yielding results competitive to state-of-the-art. Overall, we demonstrate that end-to-end max-margin training is preferred over piecewise training when combining a CNN with a CRF.

Multiatlas Segmentation Using Robust Feature-Based Registration

This paper presents a pipeline which uses a multiatlas approach for multiorgan segmentation in whole-body CT images. In order to obtain accurate registrations between the target and the atlas images, we develop an adapted feature-based method which uses organ-specific features. These features are learnt during an offline preprocessing step, and thus, the algorithm still benefits from the speed of feature-based registration methods. These feature sets are then used to obtain pairwise non-rigid transformations using RANSAC followed by a thin-plate spline refinement or NiftyReg. The fusion of the transferred atlas labels is performed using a random forest classifier, and finally, the segmentation is obtained using graph cuts with a Potts model as interaction term. Our pipeline was evaluated on 20 organs in 10 whole-body CT images at the VISCERAL Anatomy Challenge, in conjunction with the International Symposium on Biomedical Imaging, Brooklyn, New York, in April 2015. It performed best on majority of the organs, with respect to the Dice index.

Shape-aware multi-atlas segmentation

Despite of having no explicit shape model, multi-atlas approaches to image segmentation have proved to be a top-performer for several diverse datasets and imaging modalities. In this paper, we show how one can directly incorporate shape regularization into the multi-atlas framework. Unlike traditional methods, our proposed approach does not rely on label fusion on the voxel level. Instead, each registered atlas is viewed as an estimate of the position of a shape model. We evaluate and compare our method on two public benchmarks: (i) the VISCERAL Grand Challenge on multi-organ segmentation of whole-body CT images and (ii) the Hammers brain atlas of MR images for segmenting the hippocampus and the amygdala. For this wide spectrum of both easy and hard segmentation tasks, our experimental quantitative results are on par or better than state-of-the-art. More importantly, we obtain qualitatively better segmentation boundaries, for instance, preserving fine structures.