Keywords:
Pathology, Radiation safety, Physics, Dosimetry, CT, Abdomen
Authors:
D. Hough1, J. G. Fletcher1, K. Grant2, L. Yu1, J. Geske1, R. Raupach3, B. Schmidt3, T. Flohr3, C. H. McCollough1; 1Rochester, MN/US, 2Malvern, PA/US, 3Forchheim/DE
DOI:
10.1594/ecr2012/C-2621
Purpose
The purpose of this feasibility study is to investigate a prototype automatic kV selection tool for contrast-enhanced abdominopelvic CT,
evaluating its potential for radiation dose reduction and its effect on diagnostic image quality across adult patients of varying sizes and several types of abdominal CT examinations in a blinded reader study with matched control patients scanned at 120 kV.
The benefits of reduced tube kilovoltage (kV) CT have been exploited for pediatric patients for several years [1],
but have been difficult to extrapolate to an adult population mainly because of lack of tools and guidelines to predict which patients can undergo lower kV CT whilst reliably obtaining diagnostic quality images [2]. Lower kV CT images at 80 or 100 kV demonstrate greater iodine signal and greater relative iodine-related attenuation differences than similar images at 120 kV [3]. Consequently,
potential benefits of lower kV scanning at contrast-enhanced CT include reduced radiation dose with similar contrast-to-noise ratio and preservation of lesion conspicuity [4-7].
More recently,
lower kV imaging has been shown to result in improved conspicuity of liver lesions and pancreatic masses in appropriately-sized patients [8-10],
in addition to the radiation dose savings.
Additionally,
lower kV scanning may be beneficial for diagnostic purposes even when performed without dose reduction,
for example,
in patients with poor intravenous access or renal impairment in whom a smaller contrast dose or slower infusion rate may be necessary,
or in those in whom subtle attenuation differences may be diagnostically important [11,
12].
Scanning with lower kV can be a challenge in adult patients because of the increased noise and susceptibility to beam hardening and other artifacts,
as well as limitations of adult patient size,
which have only recently been addressed using rough patient size guidelines [2]. Technique charts,
which address patient body mass index or lateral width at only a specified level within the patient,
oversimplify what may be the best potential solution for every patient,
and make it difficult and time consuming to take into account tube current and pitch limitations at any particular kV for an individual CT system. The result is that the image quality at the measured portion of the patient might be appropriate,
but there may be excessive noise and artifacts elsewhere in the patient. The ideal solution would take into account patient size and tube current throughout the imaged area,
as well as the diagnostic task. Yu et al recently described a framework for considering kV reduction based on patient size and diagnostic task that takes into account iCNR as well as the acceptable limitations for image noise [13].
Determining the optimal scan settings for individual patients can be challenging given the inter-relationship between kV,
mAs,
dose,
tube current limits,
and iodine contrast and image noise.
Our objective was to determine whether the use of an automated CT kV selection tool (auto kV) can result in lower radiation dose without sacrificing image quality in contrast-enhanced abdominopelvic CT.