MENU

Click here to

×

Are you sure ?

Yes, do it No, cancel

Material-Of-Interest Based Window-Level Settings for Unenhanced Virtual Monoenergetic Images in Dual Energy CT

J Liu1*, C Ma1 , Y Zhang2 , T Lin1 , R Price1 , J Fan3 , L Chen1 , (1) Fox Chase Cancer Center, Philadelphia, Pennsylvania, (2) Tangdu Hospital, Air Force Military Medical University, Xi'an, Shaanxi, China, (3) University of South Alabama, Mobile, AL

Presentations

(Wednesday, 8/1/2018) 10:30 AM - 11:00 AM

Room: Exhibit Hall | Forum 8

Purpose: Virtual monoenergetic images (VMI) in dual-energy CT (DECT) require adjusted window/level (W/L) settings for optimal display appearance due to the variation of HU with energy. This study aims to develop a material-of-interest based method for W/L adjustments for VMIs.

Methods: A parameterized fitting of mass attenuation coefficient as a function of atomic number and photon energy was used to derive a short form of HU. The energy dependent part was derived to be integrated as a single factor Q(E). A calibration procedure was designed to determine the Q(E) factor based on measurements of HU(E) on VMIs of the Gammex-467 CT phantom on a DECT scanner (Siemens SOMATOM Definition AS). The energy independent part of HU was found to be specified by electron density and effective atomic number. The ICRU-46 human body tissue composition database was used to develop a material-of-interest model that was hypothesized as universal tissue information in real patients. Given the conventional W/L settings at 120kVp, the corresponding W/L for VMIs can be solved.

Results: The short form of HU was derived to be HU(E) = Ï?'+Q(E)*Z', with (Ï?', Z') as functions of electron density and atomic number. Piecewise linear fitting of Z' as a function of Ï?' was established using the ICRU-46 data. The material-of-interest modeled W/L at 40 to 190keV were obtained for sites including soft tissue, brain, lung, pelvis and spine, based on the conventional W/L settings. Preliminary evaluation on a brain case indicated that improved perceivable information on VMIs with the adjusted W/L could be observed.

Conclusion: The proposed method for W/L settings for VMIs provides an appropriate preset which can help reduce clinic workloads. Further adjustment of the W/L for VMIs may be needed for optimal perception based on image reader’s preferences on image contrast and noise.

Keywords

Dual-energy Imaging, Visual Perception

Taxonomy

IM- CT: Display Technology & Evaluation

Contact Email