Synthetic mask energy subtraction angiography: phantom study

Angioplasty is an interventional procedure for blood vessel stenosis in patients with vascular diseases. Digital subtraction angiography (DSA) is the gold standard imaging technique for imaging neurological and extremity vasculature. DSA enhances visibility of contrasted vessels by removing the visibility of the background structures (bone and tissue) using a pre- and post- iodine injection image subtraction. Due to the time delay between the mask and contrast image, DSA is not feasible for cardiac imaging due to significant tissue motion (lungs, heart muscles). Energy subtraction angiography (ESA) acquires two images in rapid succession using a kV-switching generator at a low and high energy, removing the need for a pre-injection image. The difference in attenuation coefficients as a function of changing energy allows for tissue suppression using a weighted log subtraction. A phantom study using an anthropomorphic chest phantom and custom 3D printed vessel was performed. The vessel was manually shifted to simulate cardiac motion, and a set of consecutive ESA images were acquired using parameters optimized in a theoretical simulation. The simulation optimized the iodine signal to noise ratio (SNR) using a stationary attenuation filter, as only a 5-15% decrease in the SNR was seen compared to using a complex rotating filter. An anti-correlated noise reduction algorithm was applied to all ESA images to compensate for the loss in iodine SNR. By evaluating the motion of the vessel across the ESA image frames, a synthetic mask image was created using an algorithm that evaluated the change in attenuation in individual pixels across the image set. The algorithm selected out the pixels that the vessel had moved out of, and reconstructs an image using only the pixels not containing the iodinated vessel. Once subtracted from an ESA image, a DSA-like image of just the vessel remains, resembling the gold standard imaging technique used for neurological imaging, in a simulated cardiac scenario. The proposed technique has the potential to yield diagnostic images of complex vasculature with fewer exposures, thereby reducing dose to the patient and procedure length.