Using the laser speckle contrast imaging and wavelet-based analyses, we investigate a latent (a "hidden") stage of the development of intracranial hemorrhages (ICHs) in newborn rats. We apply two measures based on the continuous wavelet-transform of blood flow velocity in the sagittal sinus, namely, the spectral energy in distinct frequency ranges and a multiscality degree characterizing complexity of experimental data. We show that the wavelet-based multifractal formalism reveals changes in the cerebrovascular blood flow at the development of ICH.