Processing Upper Airway Endoscopic Videos For Sleep Apnea Diagnostics.

Obstructive sleep apnea (OSA) is a common sleep disorder. Surgery is sometimes recommended, by specialists, for severe OSA. In that case, an endoscopic investigation with the patient under sedation would take place. By automating the detection of upper airway collapsing from the endoscopic video, we hope to assist doctors in their diagnostic of sleep apnea. However, completing a generic method for collapse detection is challenging in regards of the multitude of different morphologies, materials, configurations of collapses and interpretations by doctors. We established a framework for video processing and envisioned a method, composed of four intermediate modules. Our method aims to fill the VOTE table, a classification of upper airways collapses. We obtained promising results for the four modules implemented, and qualitatively evaluated the results. We also implemented a tool to display the content and tune the developed algorithms. These early results have been shown to Doctors who agreed on the good initial results.

Subject

video processing
computer vision
clinical decision support
color segmentation
obstructive sleep apnea
Drug-induced sleep endoscopy

To reference this document use:

http://resolver.tudelft.nl/uuid:ed7cedc7-4ce3-4597-a427-241b93bbff8a

Embargo date

2016-09-24

Part of collection

Student theses

Document type

master thesis

Rights

Files

PDF

Master_Thesis_Antoine_Plat.pdf

3.78 MB

Close viewer