DocumentCode
2505556
Title
Digital staining for histopathology multispectral images by the combined application of spectral enhancement and spectral transformation
Author
Bautista, Pinky A. ; Yagi, Yukako
Author_Institution
Med. Sch., Dept. of Pathology, Harvard Univ., Boston, MA, USA
fYear
2011
fDate
Aug. 30 2011-Sept. 3 2011
Firstpage
8013
Lastpage
8016
Abstract
In this paper we introduced a digital staining method for histopathology images captured with an n-band multispectral camera. The method consisted of two major processes: enhancement of the original spectral transmittance and the transformation of the enhanced transmittance to its target spectral configuration. Enhancement is accomplished by shifting the original transmittance with the scaled difference between the original transmittance and the transmittance estimated with m dominant principal component (PC) vectors; the m- PC vectors were determined from the transmittance samples of the background image. Transformation of the enhanced transmittance to the target spectral configuration was done using an nxn transformation matrix, which was derived by applying a least square method to the enhanced and target spectral training data samples of the different tissue components. Experimental results on the digital conversion of a hematoxylin and eosin (H&E) stained multispectral image to its Masson´s trichrome stained (MT) equivalent shows the viability of the method.
Keywords
least squares approximations; medical image processing; organic compounds; principal component analysis; Masson´s trichrome stained equivalent; digital staining; eosin; hematoxylin; histopathology multispectral images; least square method; principal component vectors; spectral enhancement; spectral transformation; Image color analysis; Imaging; Matrix converters; Muscles; Pathology; Training data; Vectors; Algorithms; Fibrillar Collagens; Humans; Image Enhancement; Liver; Microscopy; Staining and Labeling;
fLanguage
English
Publisher
ieee
Conference_Titel
Engineering in Medicine and Biology Society, EMBC, 2011 Annual International Conference of the IEEE
Conference_Location
Boston, MA
ISSN
1557-170X
Print_ISBN
978-1-4244-4121-1
Electronic_ISBN
1557-170X
Type
conf
DOI
10.1109/IEMBS.2011.6091976
Filename
6091976
Link To Document