Image analysis – Histogram processing – For setting a threshold
Patent
1991-04-23
1992-12-29
Mancuso, Joseph
Image analysis
Histogram processing
For setting a threshold
36441316, 36441319, 382 54, G06K 900
Patent
active
051757737
DESCRIPTION:
BRIEF SUMMARY
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a method of three-dimensional (3D) reconstruction of arborescence by labeling. The invention is primarily intended to be employed in the medical field in which the arborescences studied are angiographic arborescences. By subsequent processing of information on the reconstructed object, three-dimensional reconstruction makes it possible to present the object in any desired mode: transverse cross-sections, oblique cross-sections or even 3D display. It should be added that 3D display of 3D objects is already known The invention is essentially concerned with acquisition of geometrical data which are representative of a 3D arborescence, these data being subsequently employed in methods of visualization for displaying the arborescence. The distinctive feature of the method in accordance with the invention is that it permits reconstruction of arborescences from two-dimensional digital images in projection of the object to be reconstructed
The field of application of the invention is in particular the study of the vascular system (arterial and venous system) of any region of the human body which has a treelike structure (heart, brain, femoral artery, carotid artery, etc.). The mode of acquisition of images in projection is independent of the method. Although the invention is described in a radiology application, this latter is transposable to the case in which the images by projection are obtained by NMR, by ultrasonic insonification, and so on. Digital or analog radiology by x-rays (angiographic technique) makes it possible at the present time to obtain images which are well-suited to the practical application of the invention. The method in accordance with the invention is also applicable to any 3D filar structure other than medical.
2. Discussion of the Background
Current angiographic reconstruction techniques consist partly of techniques derived from tomodensitometric experimentation involving the use of scanners. However, the corresponding acquisitions are complicated, firstly by the need to remove from acquired images the contributions of all that does not represent the angiographic system, secondly by the fact that the flow of blood within the vessels is a phenomenon which is variable with time (and therefore calls for synchronization) and finally by the fact that the acquisition must be a three-dimensional acquisition. In order to eliminate contributions to the images by elements which are foreign to the angiographic system, it is a known practice to utilize injections of products which enhance the contrast within the capillaries. It will be borne in mind, however, that these injections cannot be repeated as often as may be desired without traumatizing the patient. The synchronization phenomenon may have the effect of increasing the duration of acquisitions. At the same time, this technique is contrary to the precautions which are necessary in order to avoid over-frequent injection of the contrast-enhancing product into a patient's blood vessels. Finally, when making use of scanner methods, three-dimensional reconstruction calls for repetition of these experiments. One of the solutions to this problem would consist in employing multi-row multidetectors in the scanners. However, this technique is essentially related to the systematic use of so-called conic projections since the x-ray source remains a point source. The algorithms of reconstruction of cross-sectional images from conic projections do not subsequently make it possible to achieve the requisite precision for permitting reconstructions. In order to overcome this disadvantage, a scanner has been designed to acquire the images of four cross-sections at the same time. The complexity of this machine is clearly multiplied by the number of simultaneous cross-sections which it is desired to acquire.
Scanner acquisition is nevertheless subject to a disadvantage: it takes place in the course of time and, in particular when it is sought to represent moving organs such as
REFERENCES:
patent: 4436684 (1984-03-01), White
patent: 4672651 (1987-06-01), Horiba et al.
patent: 4771467 (1988-09-01), Catros et al.
patent: 4843629 (1989-06-01), Mischler et al.
patent: 4945478 (1990-07-01), Merickel et al.
patent: 5034987 (1991-07-01), Fujimoto et al.
patent: 5058176 (1991-10-01), Shimazaki et al.
Systems and Computers in Japan, vol. 17, No. 1, Jan. 1986, Scripta Technica, Inc. (N.Y., US), S. Iwai et al.: "3-D reconstruction of coronary artery from cine-angiograms based on left ventricular model", pp. 26-34.
Japan Annual Reviews in Electronics, Computers & Telecommunications, Computer Science & Technologies, 1982, S. Tsuji et al.: "Knowledge-based identification of artery branches in cine-angiograms--an image understanding system which utilizes production-type knowledge", pp. 311-321.
Proceedings of the 5th International Conference on Pattern Recognition, Dec. 1-4, 1980, Fla., vol. 1, IEEE, (N.Y., US), T. Fukui et al.: "Detection and tracking of blood vessles in cine-angiograms", pp. 383-385.
IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. PAMI-6, No. 6, Nov. 1984, IEEE, G. Medioni et al.: "Matching images using linear features", pp. 675-685.
IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. PAMI-8, No. 2, Mar. 1986, IEEE, S. A. Stansfield: "ANGY: a rule-based expert system for automatic segmentation of coronary vessels from digital substracted angiograms", pp. 188-199.
Computers in Cardiology, Oct. 7-10, 1986, Boston, Mass., US, IEEE, Y. Sun et al.: "A hierarchical search algorithm for identification of coronary artery contours in digital angiograms", pp. 583-586.
Pattern Recognition, vol. 17, No. 5, 1984, Pergamon Press, Ltd, (GB), P. Kaufmann et al.: "Visual inspection using linear features", pp. 485-491.
Bouliou Alain
Coatrieux Jean-Louis
Collorec Rene
Garreau Mireille
General Electric CGR S.A.
Mancuso Joseph
LandOfFree
Method of three-dimensional reconstruction of arborescence by la does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method of three-dimensional reconstruction of arborescence by la, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method of three-dimensional reconstruction of arborescence by la will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-1893101