Date Approved

8-14-2024

Graduate Degree Type

Project

Degree Name

Medical Dosimetry (M.S.)

Degree Program

Health Professions

First Advisor

Kristen Vu

Academic Year

2023/2024

Abstract

Introduction

Accurate contouring of organs at risk (OARs) is essential in the treatment of head and neck cancer, particularly in patients with metal dental artifacts that can obscure critical structures. The ability to precisely define these OARs is crucial for minimizing radiation exposure to healthy tissues and improving patient outcomes. To address the challenges posed by metal artifacts, Metal Artifact Reduction (MAR) technology, including iterative MAR (IMAR) scanning, has been widely adopted in radiation oncology.

Methods

This retrospective study evaluated the impact of IMAR scanning and Medical Image Merge (MIM) auto-contouring software on the accuracy and efficiency of contouring OARs in twelve head and neck cancer patients with dental artifacts. Each patient underwent both IMAR and non-IMAR CT scans. Comparisons were made between manually and automatically contoured OARs, focusing on the time required and the dice similarity coefficients.

Results

Statistical analysis revealed significant differences in dice coefficients among various OARs, with the mandible achieving the highest accuracy and the lips the lowest. However, no significant difference was observed between IMAR and non-IMAR scans regarding contouring accuracy. The use of auto-contouring significantly reduced contouring time across all structures, regardless of the scan type.

Conclusion

While IMAR scanning effectively reduces metal artifacts, it does not significantly enhance contouring accuracy compared to non-IMAR scans. The study demonstrates that auto-contouring significantly improves efficiency, though contouring accuracy remains dependent on the operator’s skill. Further investigation into optimizing the integration of IMAR scanning with auto-contouring systems is warranted.

Download

Share

COinS