Babak Namazi

ORCID Identifier(s)

0000-0002-3512-3081

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

Comments

Degree granted by The University of Texas at Arlington

Download

Share

COinS