Time and event-specific deep learning for personalized risk assessment after cardiac perfusion imaging.

Standard clinical interpretation of myocardial perfusion imaging (MPI) has proven prognostic value for predicting major adverse cardiovascular events (MACE). However, personalizing predictions to a specific event type and time interval is more challenging. We demonstrate an explainable deep learning model that predicts the time-specific risk separately for all-cause death, acute coronary syndrome (ACS), and revascularization directly from MPI and 15 clinical features. We train and test the model internally using 10-fold hold-out cross-validation (n = 20,418) and externally validate it in three separate sites (n = 13,988) with MACE follow-ups for a median of 3.1 years (interquartile range [IQR]: 1.6, 3.6). We evaluate the model using the cumulative dynamic area under receiver operating curve (cAUC). The best model performance in the external cohort is observed for short-term prediction - in the first six months after the scan, mean cAUC for ACS and all-cause death reaches 0.76 (95% confidence interval [CI]: 0.75, 0.77) and 0.78 (95% CI: 0.78, 0.79), respectively. The model outperforms conventional perfusion abnormality measures at all time points for the prediction of death in both internal and external validations, with improvement increasing gradually over time. Individualized patient explanations are visualized using waterfall plots, which highlight the contribution degree and direction for each feature. This approach allows the derivation of individual event probability as a function of time as well as patient- and event-specific risk explanations that may help draw attention to modifiable risk factors. Such a method could help present post-scan risk assessments to the patient and foster shared decision-making.

Direct Risk Assessment From Myocardial Perfusion Imaging Using Explainable Deep Learning.

BACKGROUND: Myocardial perfusion imaging (MPI) is frequently used to provide risk stratification, but methods to improve the accuracy of these predictions are needed. OBJECTIVES: The authors developed an explainable deep learning (DL) model (HARD MACE [major adverse cardiac events]-DL) for the prediction of death or nonfatal myocardial infarction (MI) and validated its performance in large internal and external testing groups. METHODS: Patients undergoing single-photon emission computed tomography MPI were included, with 20,401 patients in the training and internal testing group (5 sites) and 9,019 in the external testing group (2 different sites). HARD MACE-DL uses myocardial perfusion, motion, thickening, and phase polar maps combined with age, sex, and cardiac volumes. The primary outcome was all-cause mortality or nonfatal MI. Prognostic accuracy was evaluated using area under the receiver-operating characteristic curve (AUC). RESULTS: During internal testing, patients with normal perfusion and elevated HARD MACE-DL risk were at higher risk than patients with abnormal perfusion and low HARD MACE-DL risk (annualized event rate, 2.9% vs 1.2%; P < 0.001). Patients in the highest quartile of HARD MACE-DL score had an annual rate of death or MI (4.8%) 10-fold higher than patients in the lowest quartile (0.48% per year). In external testing, the AUC for HARD MACE-DL (0.73; 95% CI: 0.71-0.75) was higher than a logistic regression model (AUC: 0.70), stress total perfusion deficit (TPD) (AUC: 0.65), and ischemic TPD (AUC: 0.63; all P < 0.01). Calibration, a measure of how well predicted risk matches actual risk, was excellent in both groups (Brier score, 0.079 for internal and 0.070 for external). CONCLUSIONS: The DL model predicts death or MI directly from MPI, by estimating patient-level risk with good calibration and improved accuracy compared with traditional quantitative approaches. The model incorporates mechanisms to explain to the physician which image regions contribute to the adverse event prediction.

Serial Assessment of Coronary Flow Reserve by Rubidium-82 Positron Emission Tomography Predicts Mortality in Heart Transplant Recipients.

OBJECTIVES: This study aimed to evaluate the long-term prognostic value of serial assessment of coronary flow reserve (CFR) by rubidium Rb 82 (82Rb) positron emission tomography (PET) in heart transplantation (HT) patients. BACKGROUND: Cardiac allograft vasculopathy is a major determinant of late mortality in HT recipients. The long-term prognostic value of serial CFR quantification by PET imaging in HT patients is unknown. METHODS: A total of 89 patients with history of HT (71% men, 7.0 ± 5.7 years post-HT, age 57 ± 11 years) scheduled for dynamic rest and stress (dipyridamole) 82Rb PET between March 1, 2008 and July 31, 2009 (PET-1) were prospectively enrolled in a single-center study. PET myocardial perfusion studies were reprocessed using U.S. Food and Drug Administration-approved software (Corridor 4DM, version 2017) for calculation of CFR. Follow-up PET (PET-2) imaging was performed in 69 patients at 1.9 ± 0.3 years following PET-1. Patients were categorized based on CFR values considering CFR ≤1.5 as low and CFR >1.5 as high CFR. RESULTS: Forty deaths occurred during the median follow-up time of 8.6 years. Low CFR at PET-1 was associated with a 2.77-fold increase in all-cause mortality (95% confidence interval [CI]: 1.34 to 5.74; p = 0.004). CFR decreased over time in patients with follow-up imaging (PET-1: 2.11 ± 0.74 vs. PET-2: 1.81 ± 0.61; p = 0.003). Twenty-five patients were reclassified based on PET-1 and PET-2 (high to low CFR: n = 18, low to high CFR: n = 7). Overall survival was similar in patients reclassified from high to low as patients with low to low CFR, whereas patients reclassified from low to high had similar survival as patients with high to high CFR. In multivariate Cox regression of patients with PET-2, higher baseline CFR (hazard ratio [HR] for a 0.73 unit (one SD) increase: 0.36, 95% CI: 0.16 to 0.82) and reduction in CFR from PET-1 to PET-2 (HR for a 0.79 unit (one SD) decrease: 1.50 to 7.84) were independent predictors of all-cause mortality. CONCLUSIONS: Serial assessment of CFR by 82Rb PET independently predicts long-term mortality in HT patients.

PET-Based Imaging of Ischemic Heart Disease.

PET-based cardiac nuclear imaging plays a large role in the management of ischemic heart disease. Compared with conventional single-photon emission CT myocardial perfusion imaging, PET provides superior accuracy in diagnosis of coronary artery disease and, with the incorporation of myocardial blood flow and coronary flow reserve, adds value in assessing prognosis for established coronary and microvascular disease. This review describes these and other uses of PET in ischemic heart disease, including assessing myocardial viability in ischemic cardiomyopathy. Developments in novel PET flow tracers and molecular imaging tools to assess atherosclerotic plaque vulnerability, vascular calcification, and vascular remodeling also are described.