This study evaluates the use of deep learning (DL) to automatically measure epicardial adipose tissue (EAT) volume and attenuation from low-dose, ungated computed tomography (CT) scans in patients undergoing hybrid myocardial perfusion imaging (MPI). The DL model, trained on 500 patients, was tested on 8281 patients, with 3511 for internal testing and 4770 for external testing. The model accurately segmented EAT within 2 seconds, compared to 15 minutes for manual annotation, and showed excellent correlation with expert annotations (Spearman correlation 0.90–0.98). During a median follow-up of 2.7 years, 565 patients experienced death or myocardial infarction (MI). Elevated EAT volume and attenuation were independently associated with an increased risk of death or MI after adjusting for relevant confounders. The study demonstrates that DL can efficiently and accurately measure EAT, providing valuable prognostic insights within the context of hybrid perfusion imaging.This study evaluates the use of deep learning (DL) to automatically measure epicardial adipose tissue (EAT) volume and attenuation from low-dose, ungated computed tomography (CT) scans in patients undergoing hybrid myocardial perfusion imaging (MPI). The DL model, trained on 500 patients, was tested on 8281 patients, with 3511 for internal testing and 4770 for external testing. The model accurately segmented EAT within 2 seconds, compared to 15 minutes for manual annotation, and showed excellent correlation with expert annotations (Spearman correlation 0.90–0.98). During a median follow-up of 2.7 years, 565 patients experienced death or myocardial infarction (MI). Elevated EAT volume and attenuation were independently associated with an increased risk of death or MI after adjusting for relevant confounders. The study demonstrates that DL can efficiently and accurately measure EAT, providing valuable prognostic insights within the context of hybrid perfusion imaging.