| Author(s): |
Adele P. Peskin; Alden A. Dima; Joe Chalfoun; |
| Title: |
Predicting Segmentation Accuracy for Biological Cell Images |
| Published: |
November 29, 2010 |
| Abstract: |
We have performed image segmentations on a very large number of images, using a wide variety of imaging conditions and cell lines, in order to study trends in the segmentation results and make predictions about segmentation accuracy. Comparing results from approximately 40,000 cells, we find a linear relationship between the highest segmentation accuracy seen for a given cell and the fraction of pixels that lie along the edge of that cell, the fraction at risk for error using any method when cells are segmented. |
| Proceedings: |
6th International Symposium on Visual Computing 2010 |
| Pages: |
pp. 549 - 560 |
| Location: |
Las Vegas, NV |
| Dates: |
November 29-December 1, 2010 |
| Keywords: |
accuracy;gradients;segmentation |
| Research Areas: |
Life Sciences Research, Bioscience & Health |
| PDF version: |
 Click here to retrieve PDF version of paper (951KB) |
