Collins, D. L., Holmes, C. J., Peters, T. M., and Evans, A. C. (1995). Automatic 3-D model-based neuroanatomical segmentation. Human Brain Mapping, 3(3):190--208.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....see the book Brain Warping edited by A.W. Toga [114] and an article by H. Lester and R.A. Simon [70] One of the most widely used registration method is the intensity based matching, which tries to align one image to another in such a way that the correlation of the image intensity is maximized [25, 26]. Alternate methods based on elastic deformation and fluid dynamics models are also available [21, 32, 44, 111, 113] Via image registration algorithms, biologically homologous points in two different im ages are identified and the mathematical transformation between these two points, called ....

....of the evolution equation should be based on an image registration method that does not assume an a priori physical model or on an empirical Bayesian framework. We will use intensity based registration algorithms that do not have explicit physical model assumptions to warp one brain to another [26, 9], but there should be further comparative studies of the different image registration methods to draw any general conclusions. It can be assumed that, in the case of morphological changes occurring in a healthy brain over a relatively short period of time, deformation occurs continuously and ....

[Article contains additional citation context not shown here]

D.L. Collins, C.J. Holmes, T.M. Peters, and A.C. Evans. Automatic 3d model-based neuroanatomical segmentation. Human Brain Mapping, 3:190 208, 1995.

....directions of algorithmic research aim at providing automatic methods to perform structure based morphometry. The first one stems directly from the iconic spatial normalisation scheme: a manual segmentation of the template is warped toward any new brain in order to obtain an automatic segmentation [5]. While this approach gives good result for stable brain areas like the deep nuclei, it is more questionable for the cortex [17] because the warping algorithms are disturbed by the high variability of the folding patterns [18, 21] Therefore, a concurent strategy for the cortex consists of linking ....

D. L. Collins, C. J. Holmes, T. M .Peters, and A. C. Evans. Automated 3D model-based neuroanatomical segmentation. Human Brain Mapping, 3:190--208, 1995.

....segmentation was actually a registration problem. Recently, Van Essen et al. 59] published work to analyze the functional and structural changes in the human cerebral cortex using surface based atlas. A discussion about this can be found in the book by Suri et al. 2] Recently, Collins et al. [60] presented a brain segmented technique that was achieved by identifying the non linear spatial transformation that best fitted maps corresponding to intensity based features between a model image and a new MRI brain volume. When completed, atlas contours defined on the model image were mapped ....

Collins, D. L., Holmes, C. J., Peters, T. M. and Evans, A. C., Automatic 3-D model-based neu- roanatomical segmentation, Human Brain Mapping, Vol. 3, No. 3, pp. 190-208, 1995.

....for a brain data set. The cross hair positions correspond to the same 3D point. Some of the structures are annotated on the 3D display. The sagittal slice is slightly zoomed in. Several registration algorithms (using both rigid and elastic matching) have been developed by di#erent groups [2, 3, 10]. Another approach is to extract anatomical knowledge from the atlas and use it to build a prior model of the input data [5] In both cases, AnatomyBrowser can serve as an aid tool in the model based segmentation loop : segmented images are used to accumulate knowledge of anatomy, which is used, ....

D.L. Collins, C.J. Holmes, T.M. Peters, and A.C. Evans. Automatic 3D model-based neuroanatomical segmentation. Human Brain Mapping, 3:190--208, 1995.

.... modeling to those offered here have been successfully applied to the case of finding image features, for example, using active shapes [7] snakes [8] 9] and landmark type methods[3] 1] 16] The deformation method as described for a single image is comparable to the work of Collins et al.[6], in which a varying resolution grid is used to register MR brain images to an atlas by balancing constraints on grid continuity and a local feature function match. A similar multiresolution approach was applied in the work by Klein et al. [10] The multiresolution approach to maximization (and the ....

D. Collins, C. Holmes, T. Peters, and A. Evans. Automatic 3-d model-based neuroanatomical segmentation. Human Brain Mapping, 3:190--208, 1995.

....more variable local features. Scalespace is the usual choice for implementing the multi resolution optimization: high scale, blurred versions of the two images are registered first, after which lower scale representations are incorporated which recapture the finer details of the original images. Collins et al. 1995 and 1996) implemented a multi scale nonlinear registration procedure on three dimensional MR brain images. Several grids of varying densities were laid down in the atlas image and the new image to be analyzed, and each grid was associated with one scale in the pre calculated scalespace of the ....

....in the middle panel, is not predicted well by the model because of the large difference in shape between the atlas and new image. Table 4.1 gives quantitative results comparing the predicted and manual segmentations. The measure used to compare these regions is the percentage overlap, used by Collins et al. 1995), defined as the number of pixels in agreement between the two segmentations divided by the total number of pixels in one region, written as a percentage. The percentage must be normalized by each region s volume in turn and results examined in case one region is entirely contained in the other. ....

Collins, D., Holmes, C., Peters, T. and Evans, A. (1995) Automatic 3-D model-based neuroanatomical segmentation. Human Brain Mapping, 3, 190--208.

.... (1) or absence (0) of the structure (Wright et al. 1995) surface extraction using a new automated method (MacDonald et al. 1996; Worsley et al. 1996c; Thompson et al. 1996; Zilles et al. 1996) and non linear deformations required to move the structure to an atlas or standardised structure (Collins et al. 1995; Kjems et al. 1996) Images derived from these different techniques can all be treated in a similar way, though of course the details differ greatly from one type of data to another. Broadly speaking, the signal to noise ratio is very small and so the signal is enhanced by two methods. The ....

Collins, D.L., Holmes, C.J., Peters, T.M., and Evans, A.C. 1995. Automatic 3-D modelbased neuroanatomical segmentation. Human Brain Mapping, 3:190-208.

....to anatomy studies, digital atlases are used for model driven segmentation. The atlas data is treated by the algorithm as a reference template, to which the input data set is registered. Several registration algorithms (using both rigid and elastic matching) have been developed by di#erent groups [2, 3, 10]. Another approach is to extract anatomical knowledge from the atlas and use it to build a prior model of the input data [5] In both cases, AnatomyBrowser can serve as an aid tool in the model based segmentation loop : segmented images are used to accumulate knowledge of anatomy, which is used, ....

D.L. Collins, C.J. Holmes, T.M. Peters, and A.C. Evans. Automatic 3D model-based neuroanatomical segmentation. Human Brain Mapping, 3:190--208, 1995.

....to find the optimal transformation such that a local image intensity similarity measure is maximized. Most methods in this class allow highly complex transformations which are normally proportional to the size of the volume. Elastic media models, viscous fluid models [4] or local smoothness models [6] are introduced as constraints to guide the non rigid spatial mapping. From these efforts, the need for non rigid transformations is by now quite clear. Note, however that these algorithms are driven by local voxel intensities. Each voxel is treated equally without taking advantage of higher level ....

D. Collins, C. Holmes, T. Peters, and A. Evans. Automatic 3D model-based neuroanatomical segmentation. Human Brain Mapping, 3(3):190--208, 1995.

....between facets can be explored while keeping a high level of conditional independence in the model. Finally, the form of p I (x) will be generalized to include several other image functions rather than the Laplacian, for instance the boundariness [7] and the correlation measure used in Collins [2]. ....

D. Collins, C. Holmes, T. Peters, and A. Evans. Automatic 3-d model-based neuroanatomical segmentation. Human Brain Mapping, 3:190-- 208, 1995.

....one subject to match the anatomy of a standard atlas brain. This then makes it possible to compare blood flow data across different subjects. This is accomplished by first aligning the brains as best as possible with linear transformations, then using 3D non linear deformations to improve the fit (Collins et al. 1995; Friston et al. 1996; Thompson Toga, 1996) This opens up the possibility of using the deformations themselves to analyse differences in brain shape. In the simplest case, we wish to test for differences in brain shape between two groups of p 1 and p 2 subjects. Let Delta ij (t) i = 1; ....

Collins, D.L., Holmes, C.J., Peters, T.M., and Evans, A.C. (1995). Automatic 3-D model-based neuroanatomical segmentation. Human Brain Mapping, 3, 190-208.

....region in the males. The average callosal shape of each group was also found, demonstrating visually the callosal shape differences between the two groups. 1 Introduction The morphological analysis of the brain from tomographic images has been a topic of active research during the past decade [1, 2, 3]. It is important for understanding the normal brain as well as for identifying specific anatomical structures affected by diseases. In this paper we propose a methodology for studying the size and shape of brain structures, based on a spatial normalization procedure [4, 5] which transforms images ....

D.L. Collins, C.J. Holmes, T.M. Peters, and A.C. Evans. Automatic 3-D model-based neuroanatomical segmentation. Human Brain Mapping, pages 190--208, 1995.

....In [Schlesinger et al. 1996] the boundaries of the objects are constrained further by using topological closed geodesic surfaces, that deform under an the potential from an image. They report good results for segmentation of major structures in the human brain. Another direction is taken in [Collins et al. 1995], where the task again is to locate objects of interest in brain images. Their method requires a completely segmented brain (a) where each volume element (voxel) has a neuroanatomical label. In order to segment another brain (b) one can recover the non linear spatial transformation field that ....

Collins, D., Holmes, C., Peters, T., and Evans, A. (1995). Automatic 3-D Model-Based Neuroanatomical Segmentation. Human Brain Mapping, 3:190--208.

....the anatomy of one subject to match the anatomy of a standard atlas brain. This then makes it possible to compare data across di erent subjects. This is accomplished by rst aligning the brains as best as possible with linear transformations, then using 3D nonlinear deformations to improve the t (Collins et al. 1995; Friston et al. 1996; Thompson Toga, 1996) This opens up the possibility of using the deformations themselves to analyse di erences in brain shape. In the simplest case, we wish to detect di erences in brain shape between two groups of p 1 and p 2 subjects. Let ij (t) i = 1; p j , ....

Collins, D.L., Holmes, C.J., Peters, T.M., and Evans, A.C. (1995). Automatic 3-D model-based neuroanatomical segmentation. Human Brain Mapping 3 190-208.

No context found.

D. L. Collins, C. J. Holmes, T. M. Peters, and A. C. Evans. Automatic 3D model-based neuroanatomical segmentation. Human Brain Mapping, 3(3):190--208, 1995.

No context found.

Collins, D. L., Holmes, C. J., Peters, T. M., and Evans, A. C. (1995). Automatic 3-D model-based neuroanatomical segmentation. Human Brain Mapping, 3(3):190--208.

No context found.

Collins, D. L., Holmes, C. J., Peters, T. M., and Evans, A. C. (1995). Automatic 3-D model-based neuroanatomical segmentation. Human Brain Mapping, 3(3):190--208.

No context found.

D.L. Collins, C.J. Holmes, T.M. Peters and A.C. Evans. Automatic 3-D Model-Based Neuroanatomical Segmentation. Human Brain Mapping, 3:190-208, 1995.

No context found.

Collins, D.L., Holmes, C.J., Peters, T.M., Evans, A.C.: Automatic 3D Model-Based Neuroanatomical Segmentation. Hum. Brain Mapp. 3, 190--208 (1995)

No context found.

D.L. Collins, C.J. Holmes, T.M. Peters and A.C. Evans. Automatic 3-D Model-Based Neuroanatomical Segmentation. Human Brain Mapping, 3:190-208, 1995.

No context found.

Collins DL, Holmes CJ, Peters TM, Evans AC. Automatic 3-D model-based neuroanatomical segmentation. Hum Brain Map 1995;3:190--208.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC