Model Library for Deformable Model-Based Segmentation of 3-D Brain MR-Images

Koikkalainen, Juha; Lötjönen, Jyrki

doi:10.1007/3-540-45786-0_67

Cited by 5 publications

(4 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The second is a study of the algorithm in context, quantify an improvement in an atlas-based technique when multiple atlases are available. This will require a method to select the most appropriate atlas for a particular subject [26,27]. A final study would involve finding correlations between clinical data and the modes discovered by atlas stratification.…”

Section: Discussionmentioning

confidence: 99%

Atlas stratification

Blezek

Miller

2007

Medical Image Analysis

View full text Add to dashboard Cite

The process of constructing an atlas typically involves selecting one individual from a sample on which to base or root the atlas. If the individual selected is far from the population mean, then the resulting atlas is biased towards this individual. This, in turn, may bias any inferences made with the atlas. Unbiased atlas construction addresses this issue by either basing the atlas on the individual which is the median of the sample or by an iterative technique whereby the atlas converges to the unknown population mean. In this paper, we explore the question of whether a single atlas is appropriate for a given sample or whether there is sufficient image based evidence from which we can infer multiple atlases, each constructed from a subset of the data. We refer to this process as atlas stratification. Essentially, we determine whether the sample, and hence the population, is multi-modal and is best represented by an atlas per mode. In this preliminary work, we use the mean shift algorithm to identify the modes of the sample and multidimensional scaling to visualize the clustering process on clinical MRI neurological image datasets.

show abstract

Section: Discussionmentioning

confidence: 99%

Atlas stratification

Blezek

Miller

2007

Medical Image Analysis

View full text Add to dashboard Cite

show abstract

“…In the model selection [12], the most similar subject to a target was chosen from the database, and its surface model was used as an a priori surface model in the non-rigid registration. The similarity was measured based on features computed for each database subject.…”

Section: Non-rigid Registration Using Ffd With Model Selectionmentioning

confidence: 99%

Individualized Geometric Model from Unorganized 3-D Points: An Application to Thorax Modeling

Koikkalainen

Lötjönen²

2003

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

Abstract. An accurate surface model is necessary in many biomedical applications. An anatomical volume image, like MR volume image, is not always available. In this paper, we compare different, mainly previously published, registration methods that can be used to generate an individual geometric model when the only information on the patient is a set of digitized points from the surface of the skin. In addition, different aspects of the 3-D point selection are studied. The comparison is performed using 22 manually segmented thorax MR volume images.

show abstract

“…By contrast, we use the image manifold to encourage energy functional convexity, and obtain optimal initializations and parameters for new images. In [11], Koikkalainen et al use a nearest neighbors approach to initialize the segmentation procedure but do not define optimal parameters for the model, nor do they make use of manifold learning to calculate distances, perform interpolation, or obtain optimal parameters for the model.…”

Section: Introductionmentioning

confidence: 99%

Is a Single Energy Functional Sufficient? Adaptive Energy Functionals and Automatic Initialization

McIntosh

Hamarneh

Medical Image Computing and Computer-Assisted Intervention – MICCAI 2007

View full text Add to dashboard Cite

Abstract. Energy functional minimization is an increasingly popular technique for image segmentation. However, it is far too commonly applied with hand-tuned parameters and initializations that have only been validated for a few images. Fixing these parameters over a set of images assumes the same parameters are ideal for each image. We highlight the effects of varying the parameters and initialization on segmentation accuracy and propose a framework for attaining improved results using image adaptive parameters and initializations. We provide an analytical definition of optimal weights for functional terms through an examination of segmentation in the context of image manifolds, where nearby images on the manifold require similar parameters and similar initializations. Our results validate that fixed parameters are insufficient in addressing the variability in real clinical data, that similar images require similar parameters, and demonstrate how these parameters correlate with the image manifold. We present significantly improved segmentations for synthetic images and a set of 470 clinical examples.

show abstract

Model Library for Deformable Model-Based Segmentation of 3-D Brain MR-Images

Cited by 5 publications

References 16 publications

Atlas stratification

Atlas stratification

Individualized Geometric Model from Unorganized 3-D Points: An Application to Thorax Modeling

Is a Single Energy Functional Sufficient? Adaptive Energy Functionals and Automatic Initialization

Contact Info

Product

Resources

About