The uncertainty decoding framework is known to improve deep neural network (DNN) based automatic speech recognition (ASR) performance in noisy environments. It operates by estimating the statistical uncertainty about the input features and propagating it to the output senone posteriors by sampling. Unfortunately, this approximate propagation scheme limits the performance improvement. In this work, we exploit the fact that uncertainty propagation can be achieved in closed form for Gaussian mixture acoustic models (GMMs). We introduce new GMM-derived (GMMD) uncertainty features for robust DNN-based acoustic model training and decoding. The GMMD features are computed as the difference between the GMM log-likelihoods obtained with vs. without uncertainty. They are concatenated with conventional acoustic features and used as inputs to the DNN. We evaluate the resulting ASR performance on the CHiME-2 and CHiME-3 datasets. The proposed features are shown to improve performance on both datasets, both for conventional decoding and for uncertainty decoding with different uncertainty estimation/propagation techniques.