Capturing the multiscale anatomical shape variability with polyaffine transformation trees

Seiler, Christof; Pennec, Xavier

doi:10.1016/j.media.2012.05.011

Cited by 35 publications

(52 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It would also be useful to see a comparison of the ICP modelling, which is relatively fast and requires very little sophisticated geometric or programming knowledge, with more sophisticated models that require specialist knowledge either to regionalize areas of interest or to implement computationally (e.g., Seiler et al 2012;Durrleman et al 2014).…”

Section: Discussionmentioning

confidence: 99%

Comparison of 3D Landmark and 3D Dense Cloud Approaches to Hominin Mandible Morphometrics Using Structure-From-Motion

Hassett

Lewis-Bale

2016

Archaeometry

View full text Add to dashboard Cite

The objective of this study is to compare landmark-based assessment of hominin mandible morphological variation with that from a high-density point cloud, using 3D models constructed from structure-from-motion (SfM) surface capture techniques. Surface models of nine hominin mandible casts were created using SfM photogrammetry. The morphology of these models was described using traditional geometric morphometrics based on identification of landmarks. This was compared to the morphological variation described by the differences between high-density point clouds, which do not rely on anatomical landmarks, using an iterative closest point algorithm. The landmark-based approach grouped the anatomically modern human and Neanderthal mandibles with reasonable success. The high-density point cloud approach also grouped these successfully, but was able to incorporate information from a specimen that was insufficiently preserved to be included in the landmark data set. This improved the accuracy of the grouping. The use of high-density point clouds from surface capture to analyse hominin mandible morphology allows for greater amounts of information to be included and offers a potential method to identify shape affinity that is as successful as landmark-based geometric morphometrics.

show abstract

Section: Discussionmentioning

confidence: 99%

Comparison of 3D Landmark and 3D Dense Cloud Approaches to Hominin Mandible Morphometrics Using Structure-From-Motion

Hassett

Lewis-Bale

2016

Archaeometry

View full text Add to dashboard Cite

show abstract

“…The method involves a cardiac motion tracking step that takes a dense displacement field computed using the LogDemons algorithm [8] and projects this to a Polyaffine space [9], subject to some cardiac-specific constraints (namely incompressibility and regional smoothing). The obtained Polyaffine parameters are then spatially and temporally aligned to a common reference frame, and the parameters for all subjects are grouped to a data tensor of [space × time × subject].…”

Section: Methodsmentioning

confidence: 99%

Descriptive and Intuitive Population-Based Cardiac Motion Analysis via Sparsity Constrained Tensor Decomposition

McLeod

Sermesant

Beerbaum

et al. 2015

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Analysing and understanding population-specific cardiac function is a challenging task due to the complex dynamics observed in both healthy and diseased subjects and the difficulty in quantitatively comparing the motion in different subjects. It was proposed to use affine parameters extracted from a Polyaffine motion model for a group of subjects to represent the 3D motion regionally over time for a group of subjects. We propose to construct from these parameters a 4-way tensor of the rotation, stretch, shear, and translation components of each affine matrix defined in an intuitive local coordinate system, stacked per region, for each affine component, over time, and for all subjects. From this tensor, Tucker decomposition can be applied with a constraint of sparsity on the core tensor in order to extract a few key, easily interpretable modes for each subject. Using this construction of a data tensor, the tensors of multiple groups can be stacked and collectively decomposed in order to compare and discriminate the motion in each group by analysing the different loadings of each combination of modes for each group. The proposed method was applied to study and compare left ventricular dynamics for a group of healthy adult subjects and a group of adults with repaired Tetralogy of Fallot.

show abstract

“…For example, in [40] multiple candidates at a landmark are generated, then the best one is selected and the others are regarded as false positives. There have been multi-scale shape models [41,42] to characterise the population variations in a more accurate and robust way. To keep our method simple, we show how a standard Gaussian shape model can be integrated with multi-scale landmark predictions We assume that all of the multi-scale predictions from LFPs are valid, but with various weights across the landmarks and scales controlled by their variances, and deduce a regulariser to obtain the ML shape with respect to the shape prior and the multi-scale landmark predictions.…”

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

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

Capturing the multiscale anatomical shape variability with polyaffine transformation trees

Cited by 35 publications

References 31 publications

Comparison of 3D Landmark and 3D Dense Cloud Approaches to Hominin Mandible Morphometrics Using Structure-From-Motion

Comparison of 3D Landmark and 3D Dense Cloud Approaches to Hominin Mandible Morphometrics Using Structure-From-Motion

Descriptive and Intuitive Population-Based Cardiac Motion Analysis via Sparsity Constrained Tensor Decomposition

Active appearance pyramids for object parametrisation and fitting

Contact Info

Product

Resources

About