Skip to Main content Skip to Navigation
Journal articles

Fast semi-automatic segmentation based on reduced basis

Abstract : This note adresses the following segmentation problem in medical imaging : minimize expert intervention for semi-automatic segmentation process. Using a reduced basis, we have an a priori knowledge of the objet we want to identify on the images, like a muscle on a CT-Scan. We just have to identify the coefficients associated to the object of interest in the reduced basis, by solving a linear system taking as input the coordinates of some selected points in the image. An example implemented in 2D is shown. This method is independent of the grayscale of the image, and can therefore be applied to all objects and images. To cite this article: Y. Maday, D. Lombardi, L. Uro, C. R. Acad. Sci. Paris, Ser. I 340 (2019).
Document type :
Journal articles
Complete list of metadata

Cited literature [16 references]  Display  Hide  Download

https://hal.inria.fr/hal-03013545
Contributor : Damiano Lombardi Connect in order to contact the contributor
Submitted on : Thursday, November 19, 2020 - 8:20:49 AM
Last modification on : Monday, December 6, 2021 - 4:30:11 PM
Long-term archiving on: : Saturday, February 20, 2021 - 6:23:44 PM

File

CRAS-220520.pdf
Files produced by the author(s)

Identifiers

Citation

Damiano Lombardi, Yvon Maday, Lydie Uro. Fast semi-automatic segmentation based on reduced basis. Comptes Rendus. Mathématique, Académie des sciences (Paris), 2020, ⟨10.5802/crmath.89⟩. ⟨hal-03013545⟩

Share

Metrics

Record views

107

Files downloads

443