Evaluation of state-of-the-art segmentation algorithms for left ventricle infarct from late Gadolinium enhancement MR images.

•We present a standardisation for evaluating segmentation algorithms for late enhancement imaging of infarct in the left ventricle.•We demonstrate that widely used fixed thresholding methods such as the standard deviation method can have inaccuracies and is user-dependent.•We provide a consensus ground truth obtained with statistical modelling on the benchmark datasets. Future algorithms can thus be benchmarked to provide a more reliable result.