Monika Christine Dadrich MD, Presenter: Speaker, Koninklijke Philips NV

Gregor Pahn DIPLPHYS, Abstract Co-Author: Nothing to Disclose

Jessica Hirsch, Abstract Co-Author: Nothing to Disclose

Johanna Laura Mayer MD, Abstract Co-Author: Nothing to Disclose

Waldemar P. Hosch MD, Abstract Co-Author: Nothing to Disclose

Hans-Ulrich Kauczor MD, Abstract Co-Author: Research Grant, Boehringer Ingelheim GmbH Research Grant, Siemens AG Research Grant, Bayer AG Speakers Bureau, Boehringer Ingelheim GmbH Speakers Bureau, Siemens AG Speakers Bureau, Novartis AG

Wolfram Stiller PhD, DIPLPHYS, Abstract Co-Author: Nothing to Disclose

Recently, iterative reconstruction algorithms (IR) have been introduced in CT, offering a new possibility for radiation dose (RD) reduction by reducing image noise resulting from CT examinations. The purpose of this study was to assess different levels of an IR algorithm (iDose, Philips Healthcare, Best, The Netherlands) in non-enhanced low-dose chest CT examinations in comparison with standard dose chest CT reconstructed with filtered backprojection (FBP).

Non-enhanced low-dose chest CT examinations (LDCT) were acquired with a tube voltage of 100kVp and a tube current-time product of 120mAs with tube-current modulation (TCM) (Group A; 20 patients), or 60mAs without TCM (Group B; 24 patients). Images were reconstructed with FBP and different levels of iDose (levels 40 and 60 with/without MFR, i.e. homogeneous noise texturing) using lung (L) and soft-tissue kernels (B). Each patient had a prior standard-dose chest CT (SDCT; 120kVp, 120mAs with TCM, FBP). CT numbers and image noise were objectively measured in different anatomic structures (lung, aorta, liver). Three independent, blinded readers assessed diagnostic image quality by subjective ranking (best to worst) of the differently reconstructed image data sets. Radiation dose parameters (CTDIw and DLP) were recorded.

Radiation exposure could be reduced by 41% (group A), and by 72% (group B), respectively. IR did not affect CT numbers while image noise could be reduced by up to ~40 %. With regard tothe assessment of subjective image quality interreader-agreement was fair to moderate (κ=0.24-0.48). Iteratively reconstructed images with iDose60 were ranked highest independent of low-dose protocol and reconstruction kernel. No difference between iDose60 with and without MFR could be observed. FBP-images of all low-dose data sets were ranked lowest.

Image noise can be reduced by IR in low-dose chest CT, thereby improving image quality compared to FBP. iDose60 is superior to iDose40, the use of MFR however doesn’t affect subjective image quality. Intra-individual comparisons between SDCT and LDCT suggest that IR enables radiation dose reduction of up to about 70%, while maintaining overall diagnostic acceptability.

IR algorithms have great potential for reducing image noise in chest CT, allowingto lower radiation exposure of chest CT examinations while preserving overall diagnostic image quality.

Dadrich, M, Pahn, G, Hirsch, J, Mayer, J, Hosch, W, Kauczor, H, Stiller, W, Iterative and Standard Filtered Back Projection Reconstruction – Comparing Image Quality of Standard and Low-dose Chest CT.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14011189.html