(Sunday, 7/12/2020) [Eastern Time (GMT-4)]
Room: AAPM ePoster Library
Purpose: To develop and evaluate the feasibility of an analytical medical image quality assessment that requires no user input, intended for the quality assessment of imaging systems during design, commissioning, and routine quality assurance.
Methods: The approach estimates image degradation factors, applies them to the known object in the image and compares the output with the test image. The comparison is executed in the histogram space, removing the spatial dependence present in Region of Interest (ROI)-based methods. As a feasibility study, planar x-ray imaging was chosen as a medical imaging technique that does not require image reconstruction and can be assumed to be quantum noise limited. The method was tested using Monte Carlo simulated planar images of a simple disk phantom.
Results: As the SNR of images decreased, the noise degradation became more significant than the degradation due to blurring in the cumulative histogram curve, and the number of data points in the fitting process decreased. The fit of the proposed cumulative histogram model to test images was affected by the inverse square increased attenuation, and the dose fall-off at the edge of the field of view. The best results were achieved with SNR values of at least 11.0 (+0.3, -0.2). For these levels, an average error in the signal and noise measurements of no more than 0.1 (+0.1, -0.1) % was obtained, while the average error in the measurement of the resolution was 0.1 (+0.2, -0.1) cycle/mm.
Conclusion: For images with an SNR of less than 10 (or less than 20 dB), the results were less promising and the error and uncertainty in noise and spatial resolution measurements were larger than conventional methods. Therefore, despite the benefits of automation offered using this approach, further work is needed to enable this approach to be applied to low dose imaging systems.
Funding Support, Disclosures, and Conflict of Interest: This work was supported by CancerCare Manitoba Foundation, the University of Manitoba and King Abdullah Scholarship Program.
Keywords
Quality Assurance, Radiography
Taxonomy
IM- X-Ray: Quality Control and Image Quality Assessment
Contact Email
