The low brightness dust emission at high Galactic latitude is of interest to study the interplay between physical processes in shaping the structure of the interstellar medium (ISM), as well as to statistically characterize dust emission as a foreground to the Cosmic Microwave Background (CMB). Progress in this avenue of research have been hampered by the difficulty of separating the dust emission from the Cosmic Infrared Background (CIB). We demonstrate that dust and CIB may be effectively separated based on their different structure on the sky and use the separation to characterize the structure of diffuse dust emission on angular scales where CIB is a significant component in terms of power. We use scattering transform statistics, the Wavelet Phase Harmonics (WPH), to perform a statistical component separation using Herschel SPIRE observations. This component separation is done only from observational data using non-Gaussian properties as a lever arm, and is done at a single 250 microns frequency. This method, that we validate on mock data, gives us access to non-Gaussian statistics of the interstellar dust and an output dust map essentially free from CIB contamination. Our statistical modelling characterizes the non-Gaussian structure of the diffuse ISM down to the smallest scales observed by Herschel. We recover the power-law shape of the dust power spectrum up to a wavenumber of 2 arcmin$^{-1}$ where the dust signal represents 2 percent of the total power. The output dust map reveals coherent structures at the smallest scales which were hidden by the CIB anisotropies. It opens new observational perspectives on the formation of structure in the diffuse ISM which we discuss with reference to past work. We have succeeded to perform a statistical separation from observational data only at a single frequency by using non-Gaussian statistics.