Room: Track 1
Purpose: To develop a system for displaying dual-energy images in real-time on an interventional C-arm and create an automatic exposure control (AEC) algorithm to guide dual-energy technique selection.
Methods: Real-time dual-energy subtraction (DES) with adjustable material cancellation factor was implemented on an interventional C-arm system equipped with a research interface for fast-kV switching. The DES processing, display and user interface were implemented in C++/OpenGL on a workstation equipped with a real-time image feed from the x-ray system. Live display of material-subtracted images was evaluated in two models: an anthropomorphic chest phantom, and a porcine model of invasive pulmonary angiography. An AEC algorithm was developed to guide dual-energy technique selection. Imaging of an iodine-equivalent reference object in different acrylic phantoms was used to establish a look-up table relating dual-energy iodine CNR to x-ray tube parameters (low kV, high kV, low ms, high ms, mA) and phantom thickness. In practice, dual-energy techniques were chosen based on estimated patient thickness and target iodine CNR value. Predicted CNR values were compared with directly measured values for several techniques in the chest phantom.
Results: Fast-kV switching was performed at 30 Hz over a range of tube voltages (55-125kV). Dual-energy fluoroscopy and angiography images were displayed in real-time at 15 fps with tissue and bone cancellation. Live DES angiography in the porcine study demonstrated reduction in diaphragm and cardiac-related motion artifacts when compared to conventional DSA. The difference between measured and target CNR in the chest phantom was 14.7% ± 2.5% (mean ± S.D.).
Conclusion: This work presents first results of a real-time system and a CNR-guided AEC developed for dual-energy imaging on an interventional C-arm. Real-time visualization of material-subtracted images enables fast evaluation of dual-energy images. Dual-energy imaging enables reduction of motion-related subtraction artifacts in invasive pulmonary angiography.
Funding Support, Disclosures, and Conflict of Interest: Financial support was provided by NIH Grant No. R21 EB023008 and funding received from Siemens Healthineers.