Viability assessment after acute myocardial infarction (MI) is important to guide revascularization. Two-dimensional strain echocardiography was shown to predict viability, but the method assumed that strain in each segment is independent of contiguous segments. The aim of this study was to test the hypotheses that segmental strain after MI is spatially correlated and that using a Bayesian approach improves the prediction of nonviable myocardium. Twenty-one subjects (mean age 58 +/- 12 years, 6 women) with MI >or=2 weeks before recruitment underwent 2-dimensional strain echocardiography and late gadolinium enhancement (LGE) cardiac magnetic resonance imaging within 48 hours of each other. The heart was divided into 16 segments, and longitudinal, radial, and circumferential strains were measured using software. Using similar segmentation, LGE was measured, and segments with >50% LGE were considered nonviable. Spearman's analyses were used to assess the spatial correlation of strain, and receiver-operating characteristic curve analysis was used to determine the prediction of nonviable myocardium without and with a Bayesian logistic spatial conditionally autoregressive (CAR) model. There was a significant spatial correlation in strain and LGE among segments, especially in the apex. Longitudinal strain was the best predictor of nonviability and was impaired in nonviable myocardium (-12.1 +/- 0.6%, -8.0 +/- 0.6%, and -4.6 +/- 1% for 0%, 1% to 50%, and >50% LGE, respectively, p <0.001). Use of the CAR model improved the area under the curve for the detection of nonviable myocardium (from 0.7 to 0.94). A CAR probabilistic score of 0.17 had 88% sensitivity and 86% specificity for detecting nonviable myocardium. In conclusion, longitudinal strain from 2-dimensional strain echocardiography can predict myocardial viability after MI, and exploiting spatial correlations in segmental strain using Bayesian CAR modeling enhances the ability of 2-dimensional strain to predict nonviable myocardium.