Imaging Science & Engineering Seminar | Abhinav K. Jha
Friday, October 10, 2025 8:30 AM to 9:30 AM
About this Event
135 N Skinker Blvd, St. Louis, MO 63112, USA
#ImagingScience, #ESEIntegrating Physics and Deep Learning for CTLESS SPECT: From Theory to Clinical Validation
Abstract: Single-photon emission computed tomography (SPECT) is a widely used clinical imaging modality but typically requires an additional X-ray CT scan for attenuation compensation. This dependence introduces several drawbacks, including increased radiation dose, higher costs, potential sources of inaccuracy, and, importantly, limited access to the necessary hybrid SPECT/CT imaging systems. In this talk, I will describe how merging concepts from information theory, physics, and deep learning enabled us to develop a solution for performing attenuation compensation directly from SPECT measurements and without the need for an additional CT scan. The presentation will trace the evolution of this approach: beginning with its theoretical foundations, moving through the development and implementation of the method, and culminating in its validation on clinically relevant tasks using anthropomorphic and, most recently, human observer studies in collaboration with multiple physicians. This journey illustrates how advances in computational imaging science can make medical imaging safer, more accessible, affordable, and accurate.
Event Details
See Who Is Interested
0 people are interested in this event