Encrypted login | home

Program Information

Initial Experience with Dose Monitoring Software: Determination of Optimal Dose Monitoring Metrics and Investigation Strategies


J Nute

J Nute*, Y Zhou , Cedars-Sinai Medical Center, Los Angeles, CA

Presentations

PO-BPC-Exhibit Hall-2 (Saturday, March 18, 2017)  Room: Exhibit Hall


Purpose: While dose monitoring software allows for automated monitoring of CT dose, the customization necessary for proper use is complex. We aimed to determine the optimal dose monitoring metrics and investigation strategies to allow for maximum oversight of CT dose with minimum man hours.

Methods: Using the Radimetrics platform, protocols are grouped under a master protocol heading which defines the scope of the protocols contained. Dose reference levels (DRLs) can be applied to each master protocol at the acquisition, exam or patient level. Available metrics include CTDIvol Head (average or maximum), CTDIvol Body (average or maximum), SSDE, and DLP. DRLs were set initially using a combination of empirical analysis and outside benchmarks. Dose alerts were then monitored over 6 months to determine optimal DRL metrics as well as identify the most useful and expedient strategies for alert investigation and resolution.

Results: While DRLs at the acquisition level were initially pursued, it was determined that they were not supported throughout the software. DRLs based on the maximum CTDIvol value within the exam provided adequate oversight of high dose exams, while DRLs based on DLP were found useful for identifying repeat acquisitions. DRLs based on both head and body CTDI values were necessary for complete oversight of neck protocols. DRLs based on SSDE were helpful in identifying modifications to base protocols, however SSDE calculations were found to be unreliable in patients with metal hardware and in exams submitted without a localizer. Dose alert investigation was most aided by visualizations of patient water-equivalent diameter (WED), localizer images, and export of similar exams for dose comparison and identification of protocol modifications.

Conclusion: A combination of several DRL metrics was found necessary to provide more comprehensive oversight of CT dose. A small range of tools were identified as the most useful in dose alert investigation.


Contact Email: