© 2015 IEEE. The lamina cribrosa is a sieve-like structure where retinal ganglion cell axons and central retinal vessels exit from the eye through the scleral canal. The lamina cribrosa has been known to play an essential role in the physiopathology of glaucoma and has been investigated as a potential location to identify early glaucomatous damage. Many researchers have been studying how the lamina cribrosa responds to intraocular pressure change, leading to axonal insults. Recently, 3D spectral domain optical coherence tomography (SD-OCT) Enhanced depth imaging (EDI), an optical imaging technique, has been proposed to improve OCT imaging of deeper retinal structures such as the choroid and the lamina cribrosa. However, the shadowing from vasculature and other reflective structures make the segmentation of the anterior lamina cribrosa surface difficult. In this paper, we present a new approach for the segmentation of the anterior lamina cribrosa surface. To deal with the complexity of the surface segmentation, we propose the use of a shape-constrained surface evolution method where the surface is refined iteratively using a non-local Markov random field based segmentation. The estimation of the model parameter is addressed using a Metropolis-Hastings algorithm. Our experiments showed a significant correlation between change in the intraocular pressure level and change in the position of the lamina cribrosa over time.