Graduation Semester and Year
2019
Language
English
Document Type
Dissertation
Degree Name
Doctor of Philosophy in Electrical Engineering
Department
Electrical Engineering
First Advisor
Venkat Devarajan
Second Advisor
Ganesh Sankaranarayanan
Abstract
Despite the advantages of minimally invasive surgeries, the indirect access and lack of the 3D field of view of the area of interest introduce complications in the procedures. Fortunately, the recorded videos from the operation offer the opportunity for intra-operative and post-operative analyses of the procedures, to improve future performance and safety. Such analysis is essential to provide the tools for evaluation and assessment of the surgeries. In this dissertation, we investigate the potential of deep learning techniques in understanding the videos captured during laparoscopic surgeries. To this end, we describe new methods for identifying the surgical instruments and the current phase of the procedure as well as the phase boundaries, which are the key components in understanding the work-flow of surgeries. Furthermore, we describe a method for analyzing and improving the safety in a laparoscopic cholecystectomy procedure by identifying the "critical view of safety" (CVS), the recognition of which is the gold standard for enhancing the safety in cholecystectomy surgery. The tools developed under the dissertation could be the essential parts of a Surgical Video Analysis System (SVAS).
Keywords
Video analysis, Laparoscopy, Deep learning, Computer vision
Disciplines
Electrical and Computer Engineering | Engineering
License
This work is licensed under a Creative Commons Attribution-NonCommercial-Share Alike 4.0 International License.
Recommended Citation
Namazi, Babak, "Towards Automated Understanding of Laparoscopic Videos" (2019). Electrical Engineering Dissertations. 334.
https://mavmatrix.uta.edu/electricaleng_dissertations/334
Comments
Degree granted by The University of Texas at Arlington