Abstract
We present a segmentation method for cell nuclei in 3D confocal images of tissue specimens, and discuss some new approaches that could substantially improve its performance. The current method combines automatic segmentation algorithms with graphic visualization and interactive classification to achieve fast and accurate segmentation. However, the edge fidelity is low for some nuclei and other nuclei are so tightly clustered that they can not be segmented. An approach based on curvature-based flow is proposed for improving edge fidelity and an approach based on the Hough transform is proposed for segmenting clusters of nuclei.
Original language | English (US) |
---|---|
Pages | 187-190 |
Number of pages | 4 |
State | Published - 1998 |
Externally published | Yes |
Event | 4th International Conference on Computer Science and Informatics, JCIS 1998, 1st International Workshop on High Performance, 1st International Workshop on Computer Vision, Pattern Recognition and Image Processing Volume 4 - Research Triangle Park, NC, United States Duration: Oct 23 1998 → Oct 28 1998 |
Other
Other | 4th International Conference on Computer Science and Informatics, JCIS 1998, 1st International Workshop on High Performance, 1st International Workshop on Computer Vision, Pattern Recognition and Image Processing Volume 4 |
---|---|
Country/Territory | United States |
City | Research Triangle Park, NC |
Period | 10/23/98 → 10/28/98 |
ASJC Scopus subject areas
- Computer Science(all)