2011 Annual International Conference of the IEEE Engineering in Medicine and Biology Society 2011
DOI: 10.1109/iembs.2011.6091985
|View full text |Cite
|
Sign up to set email alerts
|

Abstract: This paper presents the evaluation of the accuracy of an elastic registration algorithm, based on the particle filter and an optical flow process. The algorithm is applied in brain CT and MRI simulated image datasets, and MRI images from a real clinical radiotherapy case. To validate registration accuracy, standard indices for registration accuracy assessment were calculated: the dice similarity coefficient (DICE), the average symmetric distance (ASD) and the maximal distance between pixels (Dmax). The results… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

1
7
0

Year Published

2012
2012
2014
2014

Publication Types

Select...
3
2
1

Relationship

0
6

Authors

Journals

citations
Cited by 6 publications
(8 citation statements)
references
References 9 publications
1
7
0
Order By: Relevance
“…The positive results found with the No-LVM implementation follow the performance shown in [19,20], where it is stated that the use of the PF + OF method is a valuable tool for complex NRR problems. In the meantime, the negative results support the disadvantage of this methodology for multimodal cases, so that it reinforces the motivation for using LVM to achieve a multimodal registration that does not parameterise the elastic deformation space, in contrast to most of the methods reported in the literature [16][17][18].…”
Section: Clinical Imagessupporting
confidence: 81%
See 2 more Smart Citations
“…The positive results found with the No-LVM implementation follow the performance shown in [19,20], where it is stated that the use of the PF + OF method is a valuable tool for complex NRR problems. In the meantime, the negative results support the disadvantage of this methodology for multimodal cases, so that it reinforces the motivation for using LVM to achieve a multimodal registration that does not parameterise the elastic deformation space, in contrast to most of the methods reported in the literature [16][17][18].…”
Section: Clinical Imagessupporting
confidence: 81%
“…Gauss-Seidel technique [27]). For more implementation details of this strategy, the reader is referred to [19,20]. Hence, the initial estimation d 0 (r) can be refined by accumulating the displacements obtained after solving the optimisation in (2) recursively.…”
Section: Monomodal Nrr Based On Pf + Ofmentioning
confidence: 99%
See 1 more Smart Citation
“…This approach allows derivation of an optimal estimation of the registration parameters in a Bayesian fashion to overcome issues such as susceptibility to initialization and local extrema. Specifically, particle filtering was employed for rigid/affine registration of points-tosurface [17] and two point clouds [18], [19] using a point-based metric, a model-to-slice registration using a region-based metric [20], an elastic/affine registration of two images using an intensity-based metric [21], [22], and optimization of a parameterized deformable registration field [23].…”
Section: B Related Workmentioning
confidence: 99%
“…In order to apply these two algorithms (Demons and DiffDemons) for multimodal registration, an histogram matching was performed by normalizing the grayscale values of a source image based on the grayscale values of the reference one at a specified number of quantile values [16]. A most recent proposal to solve the non-rigid registration problem is based on an iterative OF framework in order to find the deformation vector field, after conducting an initial rigid registration using the PF [17]; this method has shown promising results in [18] and [19]. Nonetheless, the problem of this algorithm is its restriction to unimodal images or the necessity of an injective intensity transference function between the target and source images, which is not the case in multimodal medical image registration.…”
Section: Introductionmentioning
confidence: 99%