Optimization methods for registration of multimodal images of retina

Kubecka, Libor; Skokan, M.; Jan, J.

doi:10.1109/iembs.2003.1279818

Cited by 8 publications

(10 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The registration quality, corresponding to the transform T α , is evaluated by the criterion C. The optimal transformation T α0 transforms the floating image F into the image T α0 (F ), which contains maximal possible information about the reference image R. As shown in Ref. 18, mutual information is a robust measure of similarity, but it fails in some types of the used ophthalmic images due to falsely located extremes. Mutual information I of images R and T α0 (F ) can be computed as:…”

Section: Registrationmentioning

confidence: 99%

Optic nerve head segmentation in multimodal retinal images

et al. 2005

View full text Add to dashboard Cite

An established method for glaucoma diagnosis is the morphological analysis of the optic nerve head (ONH) by the scanning-laser-tomography (SLT). This analysis depends on prior manual outlining of the ONH. The first automated segmentation method that we developed is limited in its reliability by noise, non-uniform illumination and presence of blood vessels. Inspired by recent medical research we developed a new algorithm improving our previous method by segmenting in registered multimodal retinal images. The multimodal approach combines SLT-images with color fundus photographs (CFP). The first step of the algorithm, the registration, is based on gradient-image mutual information maximization using controlled random search as the optimization procedure. The kernel of the segmentation module consists in the anchored active contours. The initial contour is obtained from the CFP. The points the initial curve should be attracted to, the anchors, are constrained by the Hough transform applied to a morphologically processed SLT-image. The false anchors are eliminated by masking out blood vessels that are extracted in the CFP. The method was tested on 174 multimodal image pairs. The overall performance of the system yielded 89% correctly segmented ONH, qualitatively evaluated comparing the automated contours with manual ones drawn by an experienced ophthalmologist. This represents an appreciable improvement in reliability (from 74% to 89%) compared to monomodal approach. The developed method is the basis for a promising tool for glaucoma screening.

show abstract

Section: Registrationmentioning

confidence: 99%

Optic nerve head segmentation in multimodal retinal images

et al. 2005

View full text Add to dashboard Cite

show abstract

“…There are more possibilities to do that [10], one of them is to maximize a similarity criterion [6]. It can be formalized as (1) where R is the reference image and F is the floating image to be registered, which is transformed by T α to coordinates of the reference image.…”

Section: Registration By Maximization Of Mutual Informationmentioning

confidence: 99%

“…It can be formalized as (1) where R is the reference image and F is the floating image to be registered, which is transformed by T α to coordinates of the reference image. The registration quality, corresponding to the transform T α , is evaluated by the criterion C. The optimal transformation T α0 transforms the floating image F into the image T α0 (F), which contains maximal possible information about the reference image R. In [6], it has been shown that mutual information is a robust measure of similarity, but it fails in some types of used ophthalmic images due to false global extremes. Mutual information I of images R and T α (F) can be computed as:…”

Section: Registration By Maximization Of Mutual Informationmentioning

confidence: 99%

“…Thus effective use of this technique depends on finding a cooling schedule that yields good enough solutions without taking too much time. The algorithm can be formalized as on the Fig 1; for a more detailed description see [2], [6]. …”

Section: Simulated Annealingmentioning

confidence: 99%

“…The HRT is a confocal laser scanning opthalmoscope, which provides 3D image data of the human retina, from which the integral intensity image is derived [6]. The standard fundus-camera provides different information; the color photographs are used for vessel segmentation and for computing arterio-venous diameter ratio, which is important for early detection of arteriosclerosis and diabetic retinopathy.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Registration of bimodal retinal images - improving modifications

Kubecka

Jan

The 26th Annual International Conference of the IEEE Engineering in Medicine and Biology Society

Self Cite

View full text Add to dashboard Cite

Abstract-The proper optical disc segmentation in images provided by confocal laser scanning ophthalmoscope and by color fundus-camera is a necessary step in early glaucoma or arteriosclerosis detection. Fusing information from both modalities into a vector-valued image is expected to improve the segmentation reliability. The paper describes a registration of these images using optimization based on mutual information criterion function extended with gradient-image mutual information. The controlled random search (CRS) has been found more robust optimization routine than the simulated annealing (SA) while tested on a set of 174 image pairs. Finally, the multi-resolution algorithm for bimodal retinal image registration achieving the success-rate of 94% is proposed.

show abstract