Strokes are one of the primary sources of long-term disability with billions in annual direct and indirect costs to the United States healthcare system. Nearly one-third of all strokes occur in patients with clogged carotid arteries. Carotid artery imaging types include digital subtraction angiography (DSA), duplex ultrasonography (DUS), CT angiography, and MR angiography. These imaging techniques provide information on the carotid artery’s shape, localized blood flow, and plaque composition. This data provides clinicians with information on the severity of the disease, its impact on brain function, and the possibility of plaque rupture.

Additionally, artificial intelligence (AI) promises to reduce the need for ionizing radiation in medical imaging, generally and carotid plaque stratification, specifically. Neurovascular researchers are increasingly applying machine learning and artificial neural networks to carotid artery bifurcation CT data. Most AI studies for radiology involve deep learning algorithms which mimic the brain’s cognitive processes, and specifically, convolutional neural networks (CNN). CNNs enable radiologists to reduce the imaging time - and therefore the radiation dosage - imposed upon patients. 

Inventors at RPI are developing PERFICT: A Photon-counting Energy-Resolving Fine-resolution Interior Computed Tomography system. PERFICT allows for a detailed morphology and material composition of human carotid plaques. The technology’s innovations include a high-resolution, energy-discriminating, photon-counting x-ray detection subsystem, a novel high-power x-ray source with a smaller than usual focal spot, novel interior tomography algorithms for photon-counting data, and the synergistic integration of these components. The scanner will achieve spatial resolution improvement down to 70 μm, spectral separation in up to 8 energy bins, and multi-material plaque characterization over a 2-cm-diameter region of interest.