Influence of gray level discretization on radiomic feature stability for different CT scanners, tube currents and slice thicknesses: a comprehensive phantom study.

Ruben T H M LarueJanita E van Timmeren Evelyn E C de JongGiacomo FelicianiRalph T H LeijenaarWendy M J SchreursMeindert N SosefFrank H P J RaatFrans H R van der ZandeMarco DasWouter van ElmptPhilippe Lambin

Published in: Acta oncologica (Stockholm, Sweden) (2017)

CT-scanner, slice thickness and bin width affected radiomic feature values, whereas no effect of exposure was observed. Optimization of gray-level discretization to potentially improve prognostic value can be performed without compromising feature stability. Resampling images prior to feature extraction decreases the variability of radiomic features.

Keyphrases

image quality
deep learning
machine learning
dual energy
computed tomography
optical coherence tomography
contrast enhanced
convolutional neural network
positron emission tomography
neural network
magnetic resonance imaging