| 1. |
- Karlsson, Johan, et al.
(författare)
-
Shape Modeling by Optimising Description Length Using Gradients and Parameterisation Invariance
- 2012
-
Ingår i: Analysis for Science, Engineering and Beyond, The Tribute Workshop in Honour of Gunnar Sparr held in Lund, May 8-9, 2008. - Berlin, Heidelberg : Springer Berlin Heidelberg. - 9783642202353 - 9783642202360 ; 6, s. 51-91
-
Bokkapitel (övrigt vetenskapligt/konstnärligt)abstract
- In Statistical Shape Modeling, a dense correspondence between the shapes in the training set must be established. Traditionally this has been done by hand, a process that commonly requires a lot of work and is difficult, especially in 3D. In recent years there has been a lot of work on automatic construction of Shape Models. In recent papers (Davies et al., Medical Image Computing and Computer-Assisted Intervention MICCAI’2001, pp. 57–65, 2001; Davies et al., IEEE Trans. Med. Imaging. 21(5):525–537 2002; Kotcheff and Taylor, Med. Image Anal. 2:303–314 1998) Minimum Description Length, (MDL), is used to locate a dense correspondence between shapes. In this paper the gradient of the description length is derived. Using the gradient, MDL is optimised using steepest descent. The optimisation is therefore faster and experiments show that the resulting models are better. To characterise shape properties that are invariant to similarity transformations, it is first necessary to normalise with respect to the similarity transformations. This is normally done using Procrustes analysis. In this paper we propose to align shapes using the MDL criterion. The MDL based algorithm is compared to Procrustes on a number of data sets. It is concluded that there is improvement in generalisation when using MDL to align the shapes. In this paper novel theory to prevent the commonly occurring problem of clustering under correspondence optimisation is also presented. The problem is solved by calculating the covariance matrix of the shapes using a scalar product that is invariant to mutual reparameterisations. An algorithm for implementing the ideas is proposed and compared to Thodberg’s state of the art algorithm for automatic shape modeling. The suggested algorithm is more stable and the resulting models are of higher quality according to the generalisation measure and according to visual inspection of the specificity.
|
|
| 2. |
- Henriksson, Karin, et al.
(författare)
-
A pilot study of facial, cranial and brain MRI morphometry in men with schizophrenia: Part 2.
- 2006
-
Ingår i: Psychiatry Research. - : Elsevier BV. - 1872-7123 .- 0925-4927 .- 0165-1781. ; 147:2-3, s. 187-195
-
Tidskriftsartikel (refereegranskat)abstract
- This pilot study applies a new 3D morphometric MR method to test the hypothesis that men with schizophrenia (vs. controls) have deviant facial shapes and landmark relations in cranio/facialibrain (CFB) regions. This constitutes Part 2 of paired articles in this issue of Psychiatry Research: Neuroimaging, in which Part 1 presents the new method in detail. MRI coordinates from CFB landmarks of 23 patients and 15 controls were identified and then aligned with the Procrustes model, leaving shape as the only unitless geometrical information. Men with schizophrenia had significantly longer mid- and lower-facial heights, and greater lower (left) facial depth, with a tendency toward rotation along the facial midline. This supports findings from earlier anthropometric and 3D studies of the "exterior" (face). In contrast, none of the patient-control differences for the new "interior" (cranial-brain) distances reached statistical significance. These results need to be retested on a larger sample of both sexes. (c) 2006 Elsevier Ireland Ltd. All rights reserved.
|
|
| 3. |
- Ericsson, Anders, et al.
(författare)
-
An affine invariant deformable shape representation for general curves
- 2003
-
Ingår i: Proceedings of the IEEE International Conference on Computer Vision. - 0769519504 ; 2, s. 1142-1149
-
Konferensbidrag (refereegranskat)abstract
- Automatic construction of Shape Models from examples has been the focus of intense research during the last couple of years. These methods have proved to be useful for shape segmentation, tracking and shape understanding. In this paper novel theory to automate shape modelling is described. The theory is intrinsically defined for curves although curves are infinite dimensional objects. The theory is independent of parameterisation and affine transformations. We suggest a method for implementing the ideas and compare it to minimising the Description Length of the model (MDL). It turns out that the accuracy of the two methods is comparable. Both the MDL and our approach can get Stuck at local minima. Our algorithm is less computational expensive and relatively good solutions are obtained after a few iterations. The MDL is, however, better suited at fine-tuning the parameters given good initial estimates to the problem. It is shown that a combination of the two methods outperforms either on its own.
|
|
| 4. |
- Ericsson, Anders, et al.
(författare)
-
Automated interpretation of ventilation-perfusion lung scintigrams for the diagnosis of pulmonary embolism using support vector machines
- 2003
-
Ingår i: 13th Scandinavian Conference, SCIA 2003 Halmstad, Sweden, June 29 – July 2, 2003 Proceedings/Lecture Notes in Computer Science. - 0302-9743 .- 1611-3349. - 9783540406013 ; 2749, s. 415-421
-
Konferensbidrag (refereegranskat)abstract
- The purpose of this study was to develop a new completely automated method for the interpretation of ventilation-perfusion (V-P) lung scintigrams for the diagnosis of pulmonary embolism. A new way of extracting features, characteristic for pulmonary embolism is presented. These features are then used as input to a Support Vector Machine, which discriminates between pulmonary embolism or no embolism. Using a material of 509 training cases and 104 test cases, the performance of the system, measured as the area under the ROC curve, was 0.86 in the test group. It is concluded that a completely automatic method can be used for interpretation of V-P scintigrams. It is faster and more robust than a previously presented method [4,5] and the accuracy is at the same level as the the previous method. It also handles abnormalities in the lungs.
|
|
| 5. |
|
|
| 6. |
|
|
| 7. |
- Karlsson, Johan, et al.
(författare)
-
Parameterisation invariant statistical shape models
- 2004
-
Ingår i: Proceedings of the 17th International Conference on Pattern Recognition. - 0769521282 ; , s. 23-26
-
Konferensbidrag (refereegranskat)abstract
- In this paper novel theory to automate shape modelling is described. The main idea is to develop a theory that is intrinsically defined for curves, as opposed to a finite sample of points along the curves. The major problem here is to define shape variation in a way that is invariant to curve parametrisations. Instead of representing continuous curves using landmarks, the problem is treated analytically and numerical approximations are introduced at the latest stage. The problem is solved by calculating the covariance matrix of the shapes using a scalar product that is invariant to global reparametrisations. An algorithm for implementing the ideas is proposed and compared to a state of the an algorithm for automatic shape modelling. The problems with instability in earlier formulations are solved and the resulting models are of higher quality
|
|
| 8. |
- Richter, J, et al.
(författare)
-
Automated interpretation of cardiac scintigrams
- 2003
-
Ingår i: Lecture Notes in Computer Science (Image Analysis, Proceedings). - 0302-9743 .- 1611-3349. - 9783540406013 ; 2749, s. 565-570
-
Konferensbidrag (refereegranskat)abstract
- The purpose of this study was to develop an automated method for the segmentation of the heart in a 3-D cardiac scintigram. This is immediately useful for eliminating a manual step in a previous version of a decision support system. The automatic segmentation method uses a statistical 3D-model, inspired by Active Shape, which locates the base and apex automatically from a cardiac scintigram. Key features of this algorithm are that it can handle cases where there is no or very little activity in the apex and also if there axe additional parts of the heart where there is little activity. The algorithm has been tested on approximately 2000 cardiac scintigrams.
|
|
| 9. |
- Åström, Karl, et al.
(författare)
-
Automatic feature point correspondences and shape analysis with missing data and outliers using MDL
- 2007
-
Ingår i: Proceedings 15th Scandinavian Image Analysis Conference. - Berlin, Heidelberg : Springer Berlin Heidelberg. - 9783540730392 ; 4522, s. 21-30
-
Konferensbidrag (refereegranskat)abstract
- Automatic construction of shape models from examples has recently been the focus of intense research. These methods have proved to be useful for shape segmentation, tracking, recognition and shape understanding. In this paper we discuss automatic landmark selection and correspondence determination from a discrete set of landmarks, typically obtained by feature extraction. The set of landmarks may include both outliers and missing data. Our framework has a solid theoretical basis using principles of minimal description length (MDL). In order to exploit these ideas, new non-heuristic methods for (i) principal component analysis and (ii) procrustes mean are derived - as a consequence of the modelling principle. The resulting MDL criterion is optimised over both discrete and continuous decision variables. The algorithms have been implemented and tested on the problem of automatic shape extraction from feature points in image sequences.
|
|
