Dan Tong MD, Abstract Co-Author: Nothing to Disclose

Tingting Yuan, Presenter: Nothing to Disclose

To evaluate the value of spectral CT imaging in minimizing beam-hardening artifacts (BHAs) in non-enhanced cranial computed tomography (CT).

33 brain patients scanned with spectral CT imaging mode were used for the evaluation. Both conventional 140kVp polychromatic and monochromatic images (40-140keV) were generated from a single dual energy spectral CT acquisition. The image noise in the corona (as background) and medulla oblongata, cerebellar, pons and the inferior part of frontal lobes (as regions of interest) were measured in the patient study. The BHA index, defined as the square root of the squared noise difference between the region of interest and background, was calculated and compared between the polychromatic and optimal monochromatic images. CNR of selected ROI and background and image quality was also assessed and compared.

Of all patients, 73keV was associated with best CNR and lowest BHA. The image noise of background and 4 anatomic ROIs with 140kVp images and 73keV images was (8.77, 9.85, 10.11, 9.14 and 9.87) and (8.02, 8.33, 8.91, 8.23 and 8.26), respectively. The BHA index with 73keV(2.25, 3.88, 1.85 and 1.98) was better as compared to 140kVp (4.48, 5.03, 2.57 and 4.53, respectively; all p<0.001). Monochromatic images at 73keV from spectral CT increased CNR for the 4 anatomic regions by 18%, 13%, 11% and 19%, respectively.

The dual energy spectral CT imaging with the ability of generating monochromatic images substantially reduces beam hardening artifacts and improves image quality compared to the conventional CT imaging.

Monochromatic images obtained with dual energy spectral CT imaging can substantially reduce beam hardening artifacts caused by dense materials and provide more accurate CT images for diagnosis.

Tong, D, Yuan, T, Evaluation of Dual Energy Spectral CT Imaging for Beam Hardening Artifacts Correction and Image Quality Improvement.  Radiological Society of North America 2012 Scientific Assembly and Annual Meeting, November 25 - November 30, 2012 ,Chicago IL. http://archive.rsna.org/2012/12043742.html