Back to search

University of Denver

Examining Artifacts of the Watershed Segmentation

Abstract

dc:description.abstract

The watershed segmentation is an algorithm used to systematically track cell intercalary behaviors during germ band extension of the Drosophila embryo. Neighboring cells share a contracting vertical interface, called a T1, which continues contracting to a single point, a T2, and extending in the horizontal direction to create what is called a T3 interface (Fig. 1). Additionally, higher order vertices called rosettes occur when five or more cells meet at a common vertex. Simulated T2 events demonstrate that cell angle and not noise level in the image contributes to the incorrect detection of artifactual T1s in more acute angled cells and T3s for obtuse angled cells. Short T1 simulations show a systematic overestimation of T1 lengths detected by the watershed segmentation. Order three vertex simulations show central vertex displacement is biased toward the smallest angled cells. Rosette simulations of order 5 to 11 provide a working definition of rosettes in the context of the watershed segmentation in terms of short interface frequency, length, and radius of artifactual vertices.

Degree

thesis:*
Name thesis:degree_name
M. S.
Level thesis:degree_level
Masters Thesis

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Armitage, Emily Jo
Contributors dc:contributor
  • Dinah Loerke
  • Todd Blankenship
  • Michelle Knowles

Subjects

dc:subject × 5

Rights

dc:rights
Statement dc:rights
  • <p>Copyright is held by the author. User is responsible for all copyright compliance.</p>
  • Emily Jo Armitage
Language dc:language
en

Identifiers

dc:identifier.*
Repository record dc:identifier
https://digitalcommons.du.edu/etd/1716

Chain of custody

source
Harvested from
University of Denver
Base URL
digitalcommons.du.edu/do/oai/
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Armitage, Emily Jo. Examining Artifacts of the Watershed Segmentation. Masters Thesis thesis, https://digitalcommons.du.edu/etd/1716