Abstract

In modern day medicine medical images are an integral part of clinical care. They are used in almost every clinical department from diagnosis to treatment and beyond. Medical images are viewed using electronic displays of various sizes, shapes, hardware, and software. Some clinical departments, like diagnostic radiology, require electronic displays with a large dynamic range, high contrast and high resolution. Other departments do not have any requirements and will use any commercially available display in their clinical workflow.

Viewing the same medical image on different electronic displays with different hardware, software or calibration setup could influence how observers perceive and analyze these images. This occurs often when a patient is referred from diagnostic radiology to another clinical specialty department such as radiation oncology. In this case, the patient’s tumor would be diagnosed using a high-performance display while their treatment will be planned and delivered using a commercially available display.

In this dissertation, at first, an experiment was design to examine and verify the visual contrast sensitivity of observers using the two types of displays used in the clinic. Observers were tasked with detecting a modulating bar pattern using each display under different background luminance levels and ambient room illumination. The luminance response of each display was also measured for proper comparison. Second, a set of visual experiments compared the image quality of both displays in the different sections of the radiation oncology workflow. Observers were tasked with comparing medical images viewed on both displays and ranking them on a rating scale. As part of the workflow, the observers used both displays to contour tumor and healthy tissue volumes, analyze and fuse two sets of images, verify and adjust patient’s treatment position in three degrees of motion.

The results show a clear presence for the high-performance display over the commercial grade display in every step of the radiation oncology workflow. It was shown that better visualization of medical images can improve the accuracy and precision of treatment plan and treatment delivery of radiation oncology patients.

Library of Congress Subject Headings

Imaging systems in medicine; Imaging systems--Image quality; Cancer--Radiotherapy--Imaging

Publication Date

5-2021

Document Type

Dissertation

Student Type

Graduate

Degree Name

Color Science (Ph.D.)

Advisor

Richard L. Doolittle

Advisor/Committee Member

Mark D. Fairchild

Advisor/Committee Member

Michael J. Murdoch

Campus

RIT – Main Campus

Plan Codes

CLRS-PHD

Download

Share

COinS