Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.698248
Title: The effect of X-ray system settings on perception in cardiac X-ray imaging
Author: Gislason-Lee, Amber Jill
ISNI:       0000 0004 5990 1962
Awarding Body: University of Leeds
Current Institution: University of Leeds
Date of Award: 2016
Availability of Full Text:
Access through EThOS:
Full text unavailable from EThOS. Thesis embargoed until 01 Dec 2036
Access through Institution:
Abstract:
Cardiologists use coronary angiograms to diagnose and treat cardiac patients during percutaneous coronary interventions (PCI). The X-ray photon fluence used to capture an X-ray image is directly related to image noise, whereby more fluence improves image quality; however exposure to X-rays can be harmful, therefore fluence level must be optimised. Digital image processing (DIP) may enhance image quality such that it allows for lower levels of photon fluence during image capture. The effect of changes in photon fluence and DIP on perception of coronary angiograms by clinicians was determined. Computer-generated quantum noise was added to five unprocessed patient angiograms to simulate incremental reductions in photon fluence, followed by state-of-the-art DIP. The noise adding software was calibrated for clinically-relevant settings, and validated using standard objective and subjective measurements. The degraded images were viewed next to corresponding original (not degraded) images in a staircase psychophysics experiment, with and without DIP; 12 observers selected their preferred angiogram until the point of subjective equality (PSE), i.e. level of degradation at which the observer could not perceive a difference between the original and degraded images, was determined. The PSEs were 33±15% and 34±15% reductions in fluence for unprocessed and processed images respectively. In a second psychophysics experiment angiograms representing selected fluence levels, with and without DIP, were scored individually by 16 observers on a continuous scale. Scores were analysed by a statistical regression model to quantify the fluence reduction facilitated by DIP. The DIP facilitated 25% [19%, 31%] reduction in fluence compared to no processing for the same image quality score. A significant amount of image noise can be added to coronary angiograms without being perceived, indicating potential for a significant reduction in radiation exposure exclusive to that which can be facilitated by DIP. These reductions would benefit both PCI patients and personnel.
Supervisor: Treadgold, Laura A. ; Davies, Andrew G. ; Buckley, David Sponsor: Philips Healthcare
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.698248  DOI: Not available
Share: