Stability of Dynamic Radiomics Features in Cardiac MRI

BackgroundRadiomic studies on cardiac MR mainly focus on images from distinct time points rather than considering the systems dynamic nature. Recent studies have shown that radiomic features exhibit considerable variation across the cardiac cycle and that dynamic features can improve classification accuracy in downstream tasks. However, it is unclear whether the dynamic temporal evolution of radiomic features is sufficiently stable in the presence of noise. PurposeIn this work, we evaluate the stability of radiomic feature curves of cine CMR images under noise. MethodsWe extracted over 800 radiomic features from all time points of cine CMR images of 35 subjects from three cohorts with various levels of artificially added noise. The stability of feature curves is evaluated based on pairwise normalized mean squared errors, and features are ranked by their stability. ResultsFeatures exhibit a varying degree of stability, but stability is consistent across subjects. Besides generally stable and unstable features, some features are stable within the same noise level but unstable otherwise. ConclusionSome radiomic feature curves remain stable under noise while showing variability over the cardiac cycle. These features are promising candidates for improving models using dynamic rather than static feature values.