Fig. 1: Central illustration.
From: Holistic AI analysis of hybrid cardiac perfusion images for mortality prediction
Artificial intelligence (AI) model integrating fully automated multi-structure computed tomography attenuation correction (CTAC) segmentation, quantitative image analysis (radiomics), deep learning (DL)-based coronary artery calcium (CAC), and epicardial adipose tissue (EAT) in all patients undergoing myocardial perfusion imaging (MPI) single-photon emission computed tomography/computed tomography (SPECT/CT). Receiver-operating characteristics curve for all-cause mortality and area under the receiver-operating characteristic curve values of Coronary calcium (DL-CAC score), Perfusion (stress TPD), the AI CTAC model (including DL-CAC, DL-EAT, and radiomics), the AI hybrid model (combing the CTAC model with stress MPI quantitative image parameters and stress variables) and the All model (incorporating AI hybrid image features, and clinical data).
