An Automatic Brightfield Cell Segmentation Method for the Characterization of Tissue Fluidity

Presenter

Jesus Tejeda

Campus

UMass Amherst

Sponsor

Yubing Sun, Department of Mechanical and Industrial Engineering, UMass Amherst

Schedule

Session 2, 11:30 AM - 12:15 PM [Schedule by Time][Poster Grid for Time/Location]

Location

Poster Board A78, Campus Center Auditorium, Row 4 (A61-A80) [Poster Location Map]

Abstract

Tissue fluidity has been observed to be a driving factor in the wound-healing process, occurring when cells in epithelial tissue transition from a jammed to unjammed state. Tissue fluidity is reflected by the occurrence of cellular intercalation while the macroscopic tissue structure remains unchanged. To characterize when the change to a fluidic tissue occurs and to gain insight into the wound-healing process, cell area, perimeter, and intercalation frequency are important parameters to be quantified. The experimental measurements of these parameters all require cell boundary identification. In many cases, only brightfield imaging can be utilized to capture these events in extended microscopy time-lapses, limiting boundary identification methods and requiring automatic cell segmentation. This project aims to develop an automatic cell segmentation and analysis tool for brightfield microscopy that can overcome the challenges of low contrast, non-uniform lighting, optical artifacts, and densely crowded cell images. A series of algorithms were developed in MATLAB to address the challenging cell segmentation problem in phase contrast images. Images underwent a sequence of filters and transformations— boosting contrast, removing noise, and correcting light variation—to prepare them for segmentation. The popular watershed segmentation method was combined with a marker system and edge-finding method to accurately capture cell boundaries on a variety of images. Using this approach demonstrated that the previously mentioned characteristics can be successfully obtained. The success of this segmentation tool will result in the emergence of a new expedited workflow and provide insight into the role of tissue fluidity dynamics in wound healing.

Keywords

image segmentation, wound healing , digital image processing , tissue fluidity

Research Area

Engineering

SIMILAR ABSTRACTS (BY KEYWORD)