I-Corps: Translation Potential of Artificial Intelligence (AI)-Powered Medical Imaging

This I-Corps project is based on the development of artificial intelligence (AI) technologies to improve image quality and diagnostic accuracy of heart scans. Currently, a three-dimensional (3D) imaging test that uses a safe radioactive tracer and a gamma camera to show how organs and tissues work is the most widely used imaging modality to diagnose coronary artery diseases. However, attenuation in the patient's body is one of its major limitations, leading to significant false positive diagnoses. This technology uses AI-based methods to generate attenuation correction maps, dramatically reducing false positive diagnoses. This may reduce patient complication rates and downstream healthcare costs by minimizing unnecessary invasive follow-up catheterization procedures. This I-Corps project utilizes experiential learning coupled with first-hand investigation of the industry ecosystem to assess the translation potential of artificial intelligence (AI)-based single photon emission computed tomography (SPECT) attenuation correction technology. Current approaches generate attenuation corrected images directly from SPECT images without attenuation correction. This technology uses SPECT/computed tomography (CT) scanner data to train AI networks to produce CT-like attenuation maps. Then, the attenuation information is incorporated into image reconstruction algorithms to generate cardiac SPECT images with the attenuation corrected. The AI models incorporate scatter photon information and a different image reconstruction field-of-view to achieve superior accuracy of attenuation correction. This technology may lead to superior and more stable SPECT images and improve diagnosis of coronary artery diseases. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.