Deformable spatio-temporal shape models: extending active shape models to 2D+time

Hamarneh, Ghassan; Gustavsson, Tomas

doi:10.1016/j.imavis.2003.11.009

Cited by 43 publications

(13 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Efforts to extend linear shape models to the spatiotemporal domain have been made before, for example by Hamarneh and Gustavsson (2004), in whose work each sample in the dataset consists of an entire sequence of observations of the same object sampled throughout the temporal exposure window. Mitchell et al (2002) and Bosch et al (2002) employ the same strategy, extending Active Appearance Models (Cootes et al 2001) to Active Appearance Motion Models in order to segment the endocardium in echocardiograms and Magnetic Resonance image sequences, respectively.…”

Section: Statistical Shape Models and Timementioning

confidence: 99%

“…Before, inter-subject variation and dynamics were not decoupled at all Mitchell et al 2002;Hamarneh and Gustavsson 2004). Later, the application of the model was equipped with constraints to limit the firstorder derivative of shape points over time (Montagnat and Delingette 2005) or by constraining it using a distance prior (Lynch et al 2008).…”

Section: Statistical Shape Models and Timementioning

confidence: 99%

“…After landmarking and vectorization of our observations, we have CS (n L d)-dimensional shape vectors: C frames per time sequence, S subjects (with one sequence each), n L landmarks in d dimensions. As is the case in Hamarneh and Gustavsson (2004), we construct our observation matrix by 'stacking' all vectors for one subject onto each other, such that we obtain a Cn L d × S shape matrix Y:…”

Section: Representation Of Training Datamentioning

confidence: 99%

See 2 more Smart Citations

Bilinear Models for Spatio-Temporal Point Distribution Analysis

et al. 2009

View full text Add to dashboard Cite

In this work we describe the usage of bilinear statistical models as a means of factoring the shape variability into two components attributed to inter-subject variation and to the intrinsic dynamics of the human heart. We show that it is feasible to reconstruct the shape of the heart at discrete points in the cardiac cycle. Provided we are given a small number of shape instances representing the same heart at different points in the same cycle, we can use the bilinear model to establish this.Using a temporal and a spatial alignment step in the preprocessing of the shapes, around half of the reconstruction errors were on the order of the axial image resolution of 2 mm, and over 90% was within 3.5 mm. From this, we conclude that the dynamics were indeed separated from the inter-subject variability in our dataset.

show abstract

Section: Statistical Shape Models and Timementioning

confidence: 99%

Section: Statistical Shape Models and Timementioning

confidence: 99%

Section: Representation Of Training Datamentioning

confidence: 99%

See 1 more Smart Citation

Bilinear Models for Spatio-Temporal Point Distribution Analysis

et al. 2009

View full text Add to dashboard Cite

show abstract

“…An important representative of integrating BU and TD methods is Active Contour Model (ACM).It has been applied to segment the boundary in ultrasound image [8] [9]. A priori and image-based information are utilized to push the snake toward an appropriate solution at the same time.…”

Section: Introductionmentioning

confidence: 99%

Medical Ultrasound Image Segmentation Using a Hybrid Top-Down/Bottom-Up Approach

Gao

2010

2010 Chinese Conference on Pattern Recognition (CCPR)

View full text Add to dashboard Cite

This paper proposes a hybrid framework in the study of ultrasound image segmentation that combines bottom-up (BU) with top-down (TD) strategy. The BU part with high precision constructs a weight matrix combining region-and edge-cues in terms of a novel idea of designing support regions, whereas the TD part incorporates location-driven prior knowledge from the doctor and spatial coherence information from image data. Each of these approaches has a domain of applicability, so to facilitate cue combination we introduce spectral clustering method. For reasons of algorithm optimization, a crucial idea is embedded in it of Nyström approximation and manifold assumption. The results indicate that the new algorithm provides not only higher accuracy and efficiency, but also robustness of medical ultrasound image segmentation compared to the traditional NCut method.

show abstract

“…However, its drawback of loss of accuracy and low speed limits its further applications to some other areas [9,25]. Some improvements on ASM have been made in recent years [10,11,19,20]. Three kinds of improvements to the ASM are presented in this paper.…”

Section: Introductionmentioning

confidence: 99%