A Software Tool for Optimizing Daily and Monthly CT Quality Control Program Including Self-Notification and Database Management: A Pilot Deployment On a Single CT Scanner
H Hsu*, B Peng, Y Liu, Y Lee, M Wu, T Lin, P Chaudhary, R DiTusa, S Jambawalikar, Columbia University Medical Center, New York, NY
Presentations
TU-D930-IePD-F6-2 (Tuesday, 7/12/2022) 9:30 AM - 10:00 AM [Eastern Time (GMT-4)]
Exhibit Hall | Forum 6
Purpose: The aim of this study is to perform automatic analysis of daily and monthly CT quality control (QC) and to provide immediate notification of the results to technologists, managers, and medical physicists.
Methods: Daily CT QC at our institution includes measurements of CT number, noise, and uniformity; monthly QC includes slice position, table increment accuracy, slice thickness, CT number accuracy, contrast scale, resolution, low contrast detectability, and alignment light accuracy. A dockerized container including a DICOM query/retrieve server and an automatic QC analysis and notification pipeline was under development. As part of an ongoing pilot study, the analysis of daily uniformity test and the monthly slice thickness test have been implemented. A report was generated as a new DICOM file and was sent back to PACS for technologists to review. Any failed QC would automatically prompt an e-mail notification to the corresponding personnel. The software was developed using Python3 and includes libraries such as pydicom for DICOM file handling, matplotlib for data visualization, pypdf for generating reports, and pdf2dicom for adding the report to the specific QA study on PACS.
Results: As part of an ongoing pilot study, preliminary work of the software was performed on a vendor phantom dataset for uniformity test and on the ACR phantom data set for slice thickness test. Results were available within seconds after the scans. User-customizable reports were automatically generated. A QC failure notification system is a built-in feature and will send warnings to corresponding personnel. Trend analysis of the test results was also available.
Conclusion: The software developed shows feasibility of utilizing automatic phantom QC analysis for multi-scanner data and incorporates QC failure notification for managing quality assurance program. The future goal of this software is to provide an open-source analysis software for community based add-ons and updates.
Keywords
Not Applicable / None Entered.
Taxonomy
Not Applicable / None Entered.
Contact Email