Myocardial Perfusion

Myocardial perfusion imaging (MPI) using single-photon emission computed tomography (SPECT) is crucial for diagnosing cardiovascular diseases, but image quality is often hampered by low doses, limited acquisition angles, and noise. Current research focuses on improving MPI through deep learning techniques, employing architectures like U-Nets, 3D transformers, and V-Nets for tasks such as image registration, denoising, reconstruction, and segmentation of the left ventricle. These advancements aim to enhance diagnostic accuracy by improving image quality and automating analysis, ultimately leading to better patient care and more efficient workflows in cardiology.