We introduce photometric stereo reconstruction for surface analysis of mucosal tissue. Photometric stereo captures images of a surface from a single camera, but illuminated
from different light sources. From the set of images, a detailed surface reconstruction is estimated. We apply photometric stereo principles to mucosal tissue, which
raises technical challenges due to its specular appearance. Following reconstruction of the surface, we apply surface analysis techniques based on differential geometry to detect
regions that are locally spherical, suggestive of polyps. The research described in this paper is a first step towards a real-time clinical support system based on 3D reconstruction of human tissue for gastereoentrologic applications such as colonoscopy.