Baojun Li PhD, Presenter: Nothing to Disclose

Richard H. Behrman PhD, Abstract Co-Author: Nothing to Disclose

Alexander M. Norbash MD, Abstract Co-Author: Stockholder, Boston Imaging Core Laboratories, LLC

To develop an accurate, easily implementable topogram-based patient body size index as a basis for optimizing scan protocol and monitoring patient dose in X-ray CT.

Attenuation-area product (AAP) is calculated as a patient body size index from two orthogonal topograms, through nine steps: 1) detector channel data values in topograms converted to attenuation values, μ, by negative inverse log; 2) object of interest in anterior-posterior orientation topogram segmented (SAP) from the background; 3) effective size of each detector channel, bi,, within SAP calculated based on scanner geometry and detector pitch angle; 4) attenuation-area product, μibi, within SAP summed to derive initial object AAP; 5) object in lateral orientation topogram segmented (SLAT) from background, now includes object and CT table; 6) CT table excluded either with regression model of percentage table attenuation to total attenuation as function of attenuation area product, or pre-determined relationship between table height readout and detector channel location; 7) mean centroid of object within SLAT in lateral orientation topogram computed without CT table attenuation; 8) amount of elevational mis-centering estimated from mean centroid, detector isocenter, detector pitch angle and scanner geometry; 9) amount of mis-centering used to correct initial AAP based on scanner geometry. Topograms of various phantom sizes and mis-centerings acquired; attenuation area of each acquisition computed from reconstructed images as gold standard, after first converting image pixel values from HU values to μa values, with μ as attenuation coefficient and a as area per pixel for given reconstruction field-of-view.

AAP computed as described is mathematically equivalent to attenuation area calculated directly from reconstructed CT images. AAP estimation errors from the proposed approach averaged 0.583±0.297cm, or 2.72±0.523%, compared to gold standards. With off-centered phantoms average AAP estimation errors before and after proposed mis-centering correction were 10.3% and 2.1%, respectively.

AAP is an accurate patient size index basis for CT dose monitoring and protocol optimization. Mis-centering and detector channel size corrections are essential for accurately estimating patient body size from topograms, and optimizing scan protocols.

Attenuation-Area Product is an accurate method for Dose Monitoring in X-ray CT

Li, B, Behrman, R, Norbash, A, Attentuation Area Product: Developing an Accurate Body Size Method for Patient Dose Monitoring in X-ray CT and Scan Protocol Optimization.  Radiological Society of North America 2011 Scientific Assembly and Annual Meeting, November 26 - December 2, 2011 ,Chicago IL. http://archive.rsna.org/2011/11006083.html