3D Active Shape Model Segmentation with Nonlinear Shape Priors

Kirschner, Matthias; Becker, Meike; Wesarg, Stefan

doi:10.1007/978-3-642-23629-7_60

Cited by 34 publications

(27 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The parts are assumed to be conditionally independent of one another, an assumption that has demonstrated superior performance in computation and generalisation. This form of delineation readily allows integration with advanced feature searching techniques [12], and shape optimisation methods, e.g., a Bayesian inference [11] or density estimation [26]. However a deficiency is that as a coarse delineation none of current methods give consideration to unbiasedly utilising, encoding and reconstructing the entire object appearance.…”

Section: Object Class Representationmentioning

confidence: 99%

“…The shape can be either bounded by a subspace constraint [36] as in standard ASMs or optimised by a regulariser using, e.g., density estimation [26,37], a Bayesian model [11], or sparse shape composition [38,39], leading to more efficient fitting. It has been shown that utilising multiple predictions of individual landmarks can result in robust fitting.…”

Section: Shape Regularisationmentioning

confidence: 99%

See 1 more Smart Citation

Active appearance pyramids for object parametrisation and fitting

Zhang

Bhalerao

Dickenson

et al. 2016

Medical Image Analysis

View full text Add to dashboard Cite

Object class representation is one of the key problems in various medical image analysis tasks. We propose a part-based parametric appearance model we refer to as an Active Appearance Pyramid (AAP). The parts are delineated by multi-scale Local Feature Pyramids (LFPs) for superior spatial specificity and distinctiveness. An AAP models the variability within a population with local translations of multiscale parts and linear appearance variations of the assembly of the parts. It can fit and represent new instances by adjusting the shape and appearance parameters. The fitting process uses a two-step iterative strategy: local landmark searching followed by shape regularisation. We present a simultaneous local feature searching and appearance fitting algorithm based on the weighted Lucas and Kanade method. A shape regulariser is derived to calculate the maximum likelihood shape with respect to the prior and multiple landmark candidates from multi-scale LFPs, with a compact closed-form solution. We apply the 2D AAP on the modelling of variability in patients with lumbar spinal stenosis (LSS) and validate its performance on 200 studies consisting of routine axial and sagittal MRI scans. Intervertebral sagittal and parasagittal cross-sections are typically used for the diagnosis of LSS, we therefore build three AAPs on L3/4, L4/5 and L5/S1 axial cross-sections and three on parasagittal slices. Experiments show significant improvement in convergence range, robustness to local minima and segmentation precision compared with Constrained Local Models (CLMs), Active Shape Models (ASMs) and Active Appearance Models (AAMs), as well as superior performance in appearance reconstruction compared with AAMs. We also validate the performance on 3D CT volumes of hip joints from 38 studies. Compared to AAMs, AAPs achieve a higher segmentation and reconstruction precision. Moreover, AAPs have a significant improvement in efficiency, consuming about half the memory and less than 10% of the training time and 15% of the testing time.Lumbar spinal stenosis, Active appearance model,Part-based model, Active appearance pyramid 1

show abstract

Section: Object Class Representationmentioning

confidence: 99%

Section: Shape Regularisationmentioning

confidence: 99%

Active appearance pyramids for object parametrisation and fitting

Zhang

Bhalerao

Dickenson

et al. 2016

Medical Image Analysis

View full text Add to dashboard Cite

show abstract

“…Although sophisticated algorithms have been developed (e.g., using deformable M-Reps [23] or statistical shape models [12,15,17,21]), wrongly segmented regions still occur in difficult cases. Algorithm evaluation compares reference (i.e., ground truth) and automatic segmentations, which can be in form of meshes (see Fig.…”

Section: Introductionmentioning

confidence: 99%

Extended surface distance for local evaluation of 3D medical image segmentations

2015

View full text Add to dashboard Cite

The evaluation of 3D medical image segmentation quality requires a reliable detailed comparison of a reference segmentation with an automatic segmentation. It should be able to measure the quality accurately and, thus, to reveal problematic regions. While several (global) measures, providing a single quality value, are available, the only widely used local measure is the Surface Distance (i.e., point-to-surface distance). This measure, however, has significant drawbacks such as asymmetry and underestimation in distant and differently formed regions. Other available measures have limited suitability for 3D medical segmentation evaluation. We present a more reliable distance measure for assessing and analyzing local differences between automatic and reference (i.e., ground truth) 3D segmentations. We identify and overcome Surface Distance drawbacks, esp. in regions with larger dissimilarities. We evaluated our approach on four real medical image datasets. The results indicate that our measure provides more accurate local distance values

show abstract

“…Such active shape models are widely used to characterize structures in medical images [5] as well as for biomedical image segmentation [6]. There exist also variants of the technique that make use of non-linear algorithms [7].…”

Section: Introductionmentioning

confidence: 99%

Closed-form alignment of active surface models using splines

Schmitter

Unser

2017

2017 IEEE 14th International Symposium on Biomedical Imaging (ISBI 2017)

View full text Add to dashboard Cite

We propose a new formulation of the active surface model in 3D. Instead of aligning a shape dictionary through the similarity transform, we consider more flexible affine transformations and introduce an alignment method that is unbiased in the sense that it implicitly constructs a common reference shape. Our formulation is expressed in the continuous domain and we provide an algorithm to exactly implement the framework using spline-based parametric surfaces. We test our model on real 3D MRI data. A comparison with the classical active shape model shows that our method allows us to capture shape variability in a dictionary in a more precise way.Index Terms-Active surface model, 3D, parametric surfaces, splines, continuous domain Fig. 1. Real data set consisting of 14 3D MRI scans. The red meshes outline the segmented descending thoracic aorta.

show abstract

3D Active Shape Model Segmentation with Nonlinear Shape Priors

Cited by 34 publications

References 14 publications

Active appearance pyramids for object parametrisation and fitting

Active appearance pyramids for object parametrisation and fitting

Extended surface distance for local evaluation of 3D medical image segmentations

Closed-form alignment of active surface models using splines

Contact Info

Product

Resources

About