DocumentCode
2411591
Title
Detection of tissue folds in whole slide images
Author
Bautista, Pinky A. ; Yagi, Yukako
Author_Institution
Dept. of Pathology, Harvard Med. Sch., Boston, MA, USA
fYear
2009
fDate
3-6 Sept. 2009
Firstpage
3669
Lastpage
3672
Abstract
In whole slide imaging (WSI) the quality of scanned images is an interplay between the hardware specifications of the scanning device and the condition of the tissue slide itself. Tissue artifacts such as folds and bubbles have been known to affect the efficiency of a whole slide scanning system in selecting the focus points wherein the presence of the said artifacts have been found to produce blur or unfocused images. Thus, for a whole slide scanning device to produce the best image quality, even with the presence of tissue artifacts, information on the location of these artifacts should be known such that they can be avoided in the selection of the focus points. In this paper we introduced an enhancement method to emphasize and detect the location of the tissue folds from whole slide images. Results of the experiments that we conducted on various H&E stained images that were scanned using different scanners show the robustness of the method to detect tissue folds.
Keywords
biological tissues; biomedical optical imaging; image enhancement; medical image processing; H&E stained images; blur images; image artifacts; image enhancement method; image quality; scanning device; tissue artifacts; tissue fold detection; unfocused images; whole slide images; Algorithms; Artifacts; Automatic Data Processing; Automation; Color; Colorimetry; Equipment Design; Humans; Image Processing, Computer-Assisted; Light; Optics and Photonics; Pathology; Software;
fLanguage
English
Publisher
ieee
Conference_Titel
Engineering in Medicine and Biology Society, 2009. EMBC 2009. Annual International Conference of the IEEE
Conference_Location
Minneapolis, MN
ISSN
1557-170X
Print_ISBN
978-1-4244-3296-7
Electronic_ISBN
1557-170X
Type
conf
DOI
10.1109/IEMBS.2009.5334529
Filename
5334529
Link To Document