Une description multi-échelles de la structure des images de différence Tomographie par Emission de Positons
Multiscale Structure Description of Positon Emission Tomography Difference Images
OPEN ACCESS
A method is presented here which aims at analyzing Positon Emission Tomography difference images. This method is based on a explicit description of the structure of the images. Positon Emission Tomography images are used to investigate the functional organisation of the brain, looking at the cerebral blood flow . The differences between two images from the same subject lead to the changes of activity between two particular states. These differences, called "functional activations", are supposed to be specific of a particular task. The aim is then to detect functional activations while preserving individual information, unlike classical statistical methods which look mainly for the average information across several subjects . We then build a 3-dimensional linear scale-space from the original image. Objects are extracted at each level of scale in a fully-automatic way. Then they are linked across the scales to get multi-scale objects in the scale-space . A vector of measures is associated to each of these multi-scale objects in order to characterize functional activations . We present a short study to determine the relevancy of these measures and the way they can be used.
Résumé
Nous présentons ici une méthode d'analyse d'images de différence issues de la Tomographie par Emission de Positons (TEP) qui repose sur une description explicite de la structure de ces images . Les images TEP permettent, par l'intermédiaire du débit sanguin cérébral, de rendre compte de l'état fonctionnel du cerveau . En utilisant la différence entre deux images d'un même sujet, on essaye de déterminer les différences d'activité cérébrale entre deux états . Ces différences sont supposées être spécifiques d'une tâche isolée par la différence entre les deux états, et nous les appellerons «activations fonctionnelles » . L'objectif est donc de caractériser les activations fonctionnelles dans ces images de différence, tout en préservant l'information individuelle propre au sujet, ce qui n'est pas le cas des méthodes statistiques classiques, qui s'intéressent surtout à l'information moyenne sur l'ensemble des sujets. Un espace d'échelles (« scale-space ») linéaire tri-dimensionnel est d'abord construit à partir de l'image de différence originale, puis des objets sont extraits à chaque niveau d'échelle de manière entièrement automatique. Ces objets sont ensuites liés dans les échelles pour former d'autres objets dans le scale-space . Des mesures sont alors définies et associées à chacun d'eux, afin de caractériser les activations fonctionnelles . Une étude sur la pertinence des objets définis et l'utilisation possible des mesures associées est présentée.
PET imaging, Functional cerebral activation detection, linear scale-space, primal sketch
Mots clés
Imagerie TEP, Détection d'activations fonctionnelles cérébrales, multi-échelles, scale-space linéaire, primal sketch
[1] T. Lindeberg and J-O . Eklundh . Scale-space primal sketch : construction and experiments. Image and Vision Computing, vol . 10, no . 1, 1992, pp. 3-18 .
[2] C .W. Roy and C .S . Sherrington . On the regulation of the blood supply of the brain . Journal of Physiology, vol . 11, 1890, pp. 85-108 .
[3] H .C . Lou, L. Edvinsson, and E .T. MacKenzie . The concept of coupling blood flow to brain function : revision required? Annals of neurology, vol. 22, no. 3 , 1987, pp . 289-297 .
[4] B . Horwitz and O . Sporns . Neural modeling and functional neuroimaging . Human Brain Mapping, vol . 1, 1994, pp . 269-283 .
[5] J .-F. Mangin, V. Frouin, I. Bloch, B . Bendriem, and J . Lopez-Krahe . Fast nonsupervised 3D registration of PET and MR images of the brain . Journal of Cerebral Blood Flow and Metabolism, vol . 14, no . 5, 1994, pp . 749-762 .
[6] K.J. Friston . Statistical parametric mapping . In R.W. Thatcher, M . Hallet , T. Zeffiro, E .R . John, and M . Huerta, editors, Functional Neuroimaging : Technical Foundations, 1994, pp . 79-93 .
[7] P.T. Fox, M .A . Mintun, E .R. Reiman, and M .E . Raichle . Enhanced detection offocal brain response using intersubject averaging and change-distribution analysis of subtracted PET images . Journal of Cerebral Blood Flow and Metabolism, vol . 8, 1988, pp . 642-653 .
[8] K .J . Worsley, A.C . Evans, S . Marrett, and P. Neelin . A three-dimensional statistical analysis for CBF activation studies in human brain . Journal of Cerebral Blood Flow and Metabolism, vol . 12, 1992, pp . 900-918 .
[9] K.J. Worsley, S . Marrett, P. Neelin, A .C . Vandal, K.J . Friston, and A .C . Evans . A unified statistical approach for determining significant signals in images of cerebral activation . Human Brain Mapping, vol . 4, 1996, pp . 58-73 .
[10] D.L. Hunton, F.M. Miezin, R.L. Buckner, H.I. van Mier, M .E . Raichle , and S .E. Petersen . An assessment of functional-anatomical variability in neuroimaging studies . Human Brain Mapping, vol. 4, 1996, pp. 122-139 .
[11] J .B . Poline and B .M . Mazoyer. Enhanced detection in brain activation maps using a multifiltering approach . J. of Cerebral Blood Flow and Metabolism , vol . 14, 1994, pp . 639-642 .
[12] J .B . Poline and B .M . Mazoyer . Analysis of individual brain activation maps using hierarchical description and multiscale detection . IEEE Transactions on Medical Imaging, vol . 13, no. 4, 1994, pp. 702-710.
[13] K.J. Worsley, S . Marrett, P. Neelin, and A.C . Evans . Searching scale space for activation in PET images . Human Brain Mapping, vol . 4, 1996, pp . 74-90.
[14] T. Lindeberg. Scale-space theory : A basic tool for analysing structures at different scales . Journal of Applied Statistics, vol. 21, no . 2, 1994, pp . 225-270 .
[15] J .J . Koenderink and A .J. van Doorn . Dynamic shape . Biological Cybernetics , vol . 53, 1986, pp . 383-396.
[16] A .P. Witkin . Scale-space filtering . In International Joint Conference on Artificial Intelligence, 1983, pp. 1019-1023 .
[17] J .J . Koenderink . The structure of images . Biological Cybernetics, vol . 50 , 1984, pp . 363-370 .
[18] B .M. ter Haar Romeny, L.M.J . Florack, J .J . Koenderink, and M .A . Viergever. Scale space : its natural operators and differential invariants. In A.C.F. Colchester and D .J . Hawkes, editors, Information Processing in Medical Imaging, volume 511 of Lecture Notes in Computer Science, Berlin, July 1991 . Springer-Verlag, pp. 239-255 .
[19] J . Babaud, A.P. Witkin, M . Baudin, and R .O. Duda . Uniqueness of the Gaussian kernel for scale-space filtering . IEEE Trans. Pattern Analysis and Machine Intelligence, vol . 8, 1986, pp . 26-33 .
[20] T. Lindeberg . Scale-space for n-dimensional discrete signals . In Y.O . Ying , A. Toet, D . Foster, H. Heijmanns, and P. Meer, editors, Shape in Picture : Mathematical description of shape in grey-level images, volume 126 of NATO ASI F, 1994, pp. 571-590 .
[21] O . Coulon, I. Bloch, V. Frouin, and J .-F. Mangin. Analysis of PET images using 3D scale-space primal sketch . Technical Report 97D001, Ecole Nationale Supérieure des Télécommunications, 1997 .
[22] W.J. Niessen, B .M . ter Haar Romeny, and M .A. Viergever. Numerical analysis of geometry-driven diffusion equations . In Bart M . ter Haar Romeny, editor , Geometry-Driven Diffusion in Computer Vision . Kluwer Academic, 1994 , pp . 393-410 .
[23] O. Monga, R . Lengagne, and R. Deriche. Crest line extraction in volume 3d medical images . Technical Report 2338, INRIA - Projet Syntim, 1994.
[24] K. Vincken . Probabilistic multiscale image segmentation by the hyperstack . PhD thesis, Universiteit Utrecht, Faculteit der Geneeskunde, 1995 .
[25] K.L . Vincken, A .S .E. Koster, and MA . Viergever. Probabilistic segmentation of partial volume voxels . Pattern Recognition Letters, vol. 15, 1994, pp . 477-484.
[26] G . Gerig, G . Szekely, G . Israel, and M. Berger. Detection and characterization of unsharp blobs by curve evolution . In Information Processing in Medical Imaging, 1995, pp . 165-176 .
[27] T. Lindeberg . Scale-space theory in computer vision . Kluwer Academic , 1994.
[28] P.T. Jackway and R .F. Walker. Scaled gradient watersheds and cell feature extraction. In DICTA-95 Digital Image Computing : Techniques and Applications, Brisbane, Australia, December 1995 . pp . 234-239 .
[29] L . Lifshitz and S .M . Pizer. A multiresolution hierarchical approach to image segmentation based on intensity extrema . IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 12, no . 6, 1990, pp . 529-540.
[30] T. Lindeberg. Detecting salient blob-like image structures and their scales with a scale-space primal sketch : a method for focus-of-attention. International Journal of Computer Vision, vol. 11, no . 3, 1993, pp . 283-318.
[31] T. Lindeberg . Effective scale : a natural unit for measuring scale-space lifetime. IEEE Transactions on Pattern Analysis and Machine Intelligence , vol . 15, 1993, pp. 1068-1074 .
[32] L .M .J . Florack, B .M . ter Haar Romeny, J.J. Koenderink, and M.A . Viergever. Scale-space and the differential structure of images . Image and Vision Computing, vol . 10, 1992, pp . 376-388 .
[33] P. Belin, M . Zilbovicius, S . Crozier ans A . Fontaine, M .-C . Masure, and Y. Samson . Functional lateralization of auditory temporal processing in normal humans. (submitted), 1996.
[34] A .-L . Paradis, J .-F. Mangin, I . Bloch, V. Cornilleau-Pérès, E . Moulines, and D . Le Bihan . Detection of periodic signals in brain echo-planar functional images . In IEEE-EMBS, Amsterdam, 1996 .
[35] N.F. Ramsey, B .S . Kirkby, P. Van Gelderen, K .F. Berman, J .F. Duyn, J .A . Frank, V.S . Mattay, J .D . Van Horn, G . Esposito, C .T.W. Moonen, and D.R . Weinberg . Functional mapping of the human sensorimotor cortex with 3D bold fMRI correlates highly with 112 0 PET rCBF. Journal of Cerebral Blood Flow and Metabolism, vol . 16, 1996, pp . 755-764 .
[36] J-B . Poline. Détection individuelle d'activations cérébrales par tomographie d'émission de positons . PhD thesis, Université Paris VII, December 1993.