Date Thesis Awarded

4-2022

Access Type

Honors Thesis -- Access Restricted On-Campus Only

Degree Name

Bachelors of Science (BS)

Department

Physics

Advisor

Ran Yang

Committee Members

William Cooke

Rachel Frazier

Abstract

This thesis puts forward a new Convolutional Neural Network based on the U-Net Architecture, called glottisnet to identify the glottis region in the throat. Success was had in building the network, and it is very adept at recognizing the glottis in intubation manikins. Additionally, a novel approach for triggering automatic recording of endotracheal intubation videos is developed, using a facial recognition neural network. Code for the project can be found at https://github.com/c4r5son/britescope and https://github.com/c4r5son/glottisnet.

Recommended Citation

Stillman, Carson, "Development of a Video Laryngoscope with AI Assisted Epiglottis Bounding and Automatic Recording" (2022). Undergraduate Honors Theses. William & Mary. Paper 1851.
https://scholarworks.wm.edu/honorstheses/1851