A novel compound stability-fit compensation index, to be used as a quality indicator in order to accept an explanation of a neural network prediction has been proposed in the paper “Stable and explainable deep learning damage prediction for prismatic cantilever steel beam”, published recently in the Elsevier journal Computers in Industry, (Impact Factor: 3.954, Q1 red zone) by the members of the SIMT (Signal, Image and Machine Learning Team) from the West University of Timisoara. This index is computed using both the condition number and the R2 fit indicator. Extensive testing, showed the benefits of this method to completely and trustfully characterize the location and the depth of damaged beams. Link to the paper: https://www.sciencedirect.com/science/article/abs/pii/S0166361520305935