Authors: Ghasem Hajianfar, Yazdan Salimi, Mehdi Amini, Xiaotong Hong, René Nkoulou, Elnaz Jenabi, Zahra Mansouri, Atena Aghaee, Soroush Bagheri, Amirhossein Sanaat, Ahmad Bitarafan-Rajabi, Hossein Arabi, Isaac Shiri, Habib Zaidi
Summary: A new deep learning method delivers CT-free attenuation correction for cardiac SPECT that physicians can’t tell apart from the real thing, combining multiple reconstruction inputs with dual-domain training to beat existing approaches.