Preterm birth is associated with variable degrees of brain injury and adverse neurodevelopmental outcomes. Neuroimaging has been investigated as a predictor of outcome in this population. Head ultrasound allows for rapid bedside evaluation of the neonatal brain for early intraventricular hemorrhage surveillance and later detection of periventricular leukomalacia. Computed tomography can provide excellent views for bones, hemorrhage, extra-axial space, and the ventricles but is rarely used for prognostic purposes. Magnetic resonance imaging allows for high-resolution images of brain structures, differentiation of white and gray matter, visualization of the brain stem and posterior fossa, and getting additional physiological information with specialized sequences. Though controversial, the use of magnetic resonance imaging, at term equivalent, as a predictor of later outcome in preterm infants has been increasing and has been advocated by some as a standard practice. In this article, we review and contrast the use of these various imaging modalities in predicting neurodevelopmental outcome of premature infants.