An Easy-To-Use Tool for Annual MRI Performance Evaluation Using the ACR Large Phantom
J Storrs1*, H Khosravi1, A Fatemi1,2, (1) University of Mississippi Medical Center, Jackson, MS, (2) Merit Health Cancer Center, SpinTecx.LLC, Madison, MS
Presentations
PO-GePV-I-6 (Sunday, 7/25/2021) [Eastern Time (GMT-4)]
Purpose: The ACR MRI Accreditation Program uses approved ACR phantoms for performance evaluation and quality control. A guidance documents standardize analysis by specifying seven quantitative procedures for evaluating scanner performance. These are evaluated at least annually by a qualified physicist, while three are performed weekly by technologists. The purpose of this work was to develop a fast, interactive tool to quickly assess ACR QC scans of the Large Phantom for use during annual evaluation and troubleshooting.
Methods: An interactive program for inspecting scans of the ACR large phantom was developed using Matlab. The tool accepts ACR T1 sagittal, T1 axial and PD+T2 axial series (or site axial series) and evaluates them using procedures as described in the ACR guidance document. For steps that rely on subjective input (windowing, ROI selection, etc), the most pessimistic (i.e. least beneficial toward action limits) choice is selected. Results are displayed as an interactive annotated stack. Measurements are overlaid (lines, ROIs, boxes, text) on each slice with color-coding to indicate action limit outcomes (green=pass, yellow=warning, red=fail). Clicking on a measurement adjusts the presentation to best visualize the measurement and opens relevant graphs for inspection (e.g. 1D profiles). Images and results can then be saved and exported for reporting.
Results: The tool was developed using archived annual and weekly QC scans from six scanners. The tool was then installed on a laptop and used during annual performance evaluation (5 scanners) and acceptance testing (2 scanners). Images were transferred to the laptop using USB export for evaluation. Results were rapid and because they follow the manual procedures, were easy to inspect interpret and verify in real-time.
Conclusion: An interactive tool was developed for evaluation of the ACR large phantom. The tool closely follows ACR guidance which aids rapid interpretation and understanding.
Keywords
Phantoms, Quality Control, Software
Taxonomy
IM- MRI : Quality Control and Image Quality Assessment
Contact Email