Click here to

Session: Advanced Imaging Applications for Radiotheranostics [Return to Session]

Use of Automation in Image Quality Analysis in PET

T Moretti*, S Leon, C Schaeffer, M Arreola, University of Florida, Gainesville, FL

Presentations

SU-F-202-5 (Sunday, 7/10/2022) 2:00 PM - 3:00 PM [Eastern Time (GMT-4)]

Room 202

Purpose: Certain recommended physics tests in PET are prohibitively time-consuming to perform with manual data analysis. To address this, custom software was created to perform this analysis for common physics tests. Additionally, the incorporation of metrics not originally included in the test protocols which may still be insightful to system performance were also added to the software, with the goal of comparing different imaging protocols within a system or different imaging systems.

Methods: Software was developed in MATLAB to automatically analyze DICOM image data for 3 different phantoms: a NEMA image quality phantom, suspended point sources, and a uniformity phantom. From these images the following metrics were collected: percent contrast, percent background variability, percent relative error from scatter and attenuation corrections, recovery coefficients, contrast-to-noise ratio, noise power spectrum, modulation transfer function, spatial resolution defined as the full-width at half maximum of the point sources, and integral uniformity. An executable version of the MATLAB software was developed which only requires download of a runtime license rather than a purchased MATLAB license.

Results: Data obtained from three scanners, a Siemens Biograph mCT Flow, a Siemens Biograph TruePoint, and a GE Discovery 610 were successfully analyzed for all three phantoms. Run times for all tests were less than one minute. An additional feature of the software allows for comparison between different image datasets for some of the advanced metrics. Generated results are easily incorporated into annual physics reports.

Conclusion: Software can be used to alleviate difficulties in analysis prohibiting determination of more advanced image quality metrics in PET. Different systems or reconstruction protocols can easily be compared using metrics calculated by the software.

Keywords

MTF, Noise Power Spectrum, Image Analysis

Taxonomy

IM- PET : Quality Control and Image Quality Assessment

Contact Email

Share: