By Erik L. Ridley, AuntMinnie staff writer

November 8, 2018 --

Thursday, November 29 | 11:20 a.m.-11:30 a.m. | SSQ18-06 | Room N228
Retrospective quantitative assessment of radiation dose on CT studies can yield valuable insights for improving consistency and patient safety, researchers from Duke University have found.

Most large medical institutions employ multiple CT systems of different vendors and models, and the imaging protocols across these scanners aren't necessarily consistent or applied consistently, according to presenter Aiping Ding, PhD. What's more, each imaging protocol has different dose requirements due to the variations in anatomical region, patient size, and imaging task.

There is a need to standardize dose -- as well as image quality -- across systems, he said. To create a practical approach for retrospectively assessing variation in abdominopelvic CT studies, the researchers developed an in-house informatics system that automatically extracts protocol information, patient size, radiation dose, and in vivo noise magnitude within images.

While dose monitoring can provide clinically relevant information, it's equally important to know how to use the data; collecting data without having a process to convert useful information into knowledge for improved patient care would be inconsistent with the expectations of stewardship and value-based practices, Ding said.

"In this project, we've developed a number of analytical techniques toward that end, which include dose trends and variations within patient cohorts; inconsistencies across systems, operators, and patients of different sizes; and identifying outliers toward optimization of CT practice," Ding told "Ascertaining these dose (and image quality) trends enables one to adjust the protocol parameters associated with different systems according to the patient size so that patient dose and, thus, image quality could meet certain target requirements, in effect making patient care (quality and safety) more consistent and less system-dependent."