#49: Using Deep Learning Software to Analyze Zebrafish Eye Movements and Assess Optic Nerve Regeneration — Presentation Detail — Celebration of Scholars

Instructions

Student presentations must have a faculty sponsor.

Abstracts must include a title and a description of the research, scholarship, or creative work. The description should be 150-225 words in length and constructed in a format or style appropriate for the presenter’s discipline.

The following points should be addressed within the selected format or style for the abstract:

A clear statement of the problem or question you pursued, or the scholarly goal or creative theme achieved in your work.
A brief comment about the significance or uniqueness of the work.
A clear description of the methods used to achieve the purpose or goals for the work.
A statement of the conclusions, results, outcomes, or recommendations, or if the work is still in progress, the results you expect to report at the event.

Presenter photographs should be head and shoulder shots comparable to passport photos.

Additional Information

More information is available at carthage.edu/celebration-scholars/. The following are members of the Research, Scholarship, and Creativity Committee who are eager to listen to ideas and answer questions:

Jun Wang
Kim Instenes
John Kirk
Nora Nickels
Andrew Pustina
James Ripley

#49: Using Deep Learning Software to Analyze Zebrafish Eye Movements and Assess Optic Nerve Regeneration

Name: Ashley Gorman
Major: Data Science
Hometown: Mount Pleasant, WI
Faculty Sponsor: Steven Henle
Other Sponsors:
Type of research: Independent research

Name: Claire Wulf
Major: Computer Science/Neuroscience
Hometown: Hudson, WI
Faculty Sponsor: Steven Henle
Other Sponsors:
Type of research: Independent research

Abstract

Zebrafish, unlike humans, can regenerate their optic nerves after injury. However, it is difficult to measure how well zebrafish have recovered from an optic nerve injury, and we are developing improved methods to do this. To better understand the regenerative process, we can measure the optokinetic response (OKR), a reflex that allows organisms to smoothly track movement and provides a way for us to measure whether vision has been functionally recovered after injury. We use the OKR Vision Test, which entails placing the zebrafish in a rotating 3D printed drum with striped bars and capturing video of the fish’s changing eye angles. Increasing the speed of the bars and measuring the eye movements can be used to assess how well the fish see. Previously, ocular angles for adult zebrafish had to be manually measured in FIJI and run through an R program to create a graph. We are currently working to create a more efficient method of data analysis for zebrafish by using the 3D pose estimation software DeepLabCut. Within minutes, DeepLabCut allows the user to upload videos, train a deep neural network to track individual features, and generate graphs that illustrate the rate of movement. If we better our understanding of neuronal regeneration in zebrafish, we may eventually be able to apply these findings to humans and improve the treatment of nervous system injuries.

Poster file