Bernhard Schmidt PhD, Presenter: Employee, Siemens AG

Rainer Raupach PhD, Abstract Co-Author: Employee, Siemens AG

Martin Ulrich Sedlmair MS, Abstract Co-Author: Employee, Siemens AG

Katharine Grant PhD, Abstract Co-Author: Employee, Siemens AG

Thomas G. Flohr PhD, Abstract Co-Author: Employee, Siemens AG

Previous studies have shown that in contrast enhanced CT (CTA) dose can be reduced substantially by adapting tube voltage (kV) to the attenuation of the patient. To maintain a certain image quality, typically the iodine signal to noise ratio is kept constant. Since the iodine signal increases strongly for lower kVs, an increase in image noise; and thus a decrease in dose, is acceptable to maintain image quality. However, in the case of accidental findings in anatomical areas without iodine uptake (which do not benefit from lower kVs), diagnosis may be impacted due to the increased noise. The purpose of our study is to investigate if the application of an iterative reconstruction (IR) technique after kV adaptation, could compensate for the loss in image quality in those areas.

We acquired data from anthropomorphical phantoms of different size representing a slim (28cmx20cm), average (35cmx25cm) and obese (40cmx30cm) patient using a Definition Flash (Siemens, Germany). Each of the phantoms was equipped with inserts representing soft contrast lesions with and without iodine. Data were acquired with standard scan parameters (120kV, 110 mAs, 128x0.6mm) with and without automatic kV adaptation (CAREkV: CTA preset). Data were reconstructed with regular reconstruction (B30 kernel) and using IR techniques (I30 kernel). Noise, contrast, and dose were evaluated.

Using automatic kV adaptation, kV was reduced for all phantom sizes, leading to a dose reduction by 53%, 38% and 37% for the small, medium and large phantom respectively. The reduction in kVs resulted in a signal increase by about 60%, noise increased by 56%, 44% and 42% for the small, medium and large phantom respectively, which led to reduction of low contrast detectability in lesions without contrast agent. However, when applying IR, image noise was reduced for all phantoms back to the initial level (difference to standard <5%).

Our results indicate that automatic kV adaptation allows for a dose reduction of up to 50% without compromising the iodine signal to noise ratio. To maintain noise level in areas of non-enhanced structures, the initial noise level can be preserved by applying IR techniques.

Combining iterative reconstruction approaches with an automatic selection of kVs is an efficient approach to reduce patient dose while maintaining image quality.

Schmidt, B, Raupach, R, Sedlmair, M, Grant, K, Flohr, T, Assessment of the Potential for Dose Reduction by Combining Iterative Reconstruction Techniques and Automatic kV Selection.  Radiological Society of North America 2011 Scientific Assembly and Annual Meeting, November 26 - December 2, 2011 ,Chicago IL. http://archive.rsna.org/2011/11034568.html