3D tensor-based blind multi-spectral image decomposition for tumor demarcation (CROSBI ID 564154)
Prilog sa skupa u zborniku | izvorni znanstveni rad | međunarodna recenzija
Podaci o odgovornosti
Kopriva, Ivica ; Peršin, Antun
engleski
3D tensor-based blind multi-spectral image decomposition for tumor demarcation
Blind decomposition of multi-spectral fluorescent image for tumor demarcation is formulated exploiting tensorial structure of the image. First contribution of the paper is identification of the matrix of spectral responses and 3D tensor of spatial distributions of the materials present in the image from Tucker3 or PARAFAC models of 3D image tensor. Second contribution of the paper is clustering based estimation of the number of the materials present in the image as well as matrix of their spectral profiles. 3D tensor of the spatial distributions of the materials is recovered through 3-mode multiplication of the multi-spectral image tensor and inverse of the matrix of spectral profiles. Tensor representation of the multi-spectral image preserves its local spatial structure that is lost, due to vectorization process, when matrix factorization-based decomposition methods (such as non-negative matrix factorization and independent component analysis) are used. Superior performance of the tensor-based image decomposition over matrix factorization-based decompositions is demonstrated on experimental red-green-blue (RGB) image with known ground truth as well as on RGB fluorescent images of the skin tumor (basal cell carcinoma).
3D tensor ; fluorescent multi-spectral image ; data clustering ; tumor demarcation
nije evidentirano
nije evidentirano
nije evidentirano
nije evidentirano
nije evidentirano
nije evidentirano
Podaci o prilogu
76231W-1-76231W-8.
2010.
objavljeno
10.1117/12.839568
Podaci o matičnoj publikaciji
Proc. of SPIE Vol. 7623
Dawant, Benoit M ; Haynord, David R
Bellingham (WA): SPIE
978-0-8149-8031-6
Podaci o skupu
Medical Imaging 2010: Image Processing
poster
13.02.2010-18.02.2010
San Diego (CA), Sjedinjene Američke Države
Povezanost rada
Matematika, Računarstvo, Kliničke medicinske znanosti