EDGE-DETECTION, 3-DIMENSIONAL CELL BOUNDARY RECONSTRUCTION AND VOLUMEAND SURFACE-AREA ESTIMATION FROM DIFFERENTIAL INTERFERENCE CONTRAST IMAGES

Three-dimensional (3-D) cell morphology is important for the understan ding of cell function and can by quantified in terms of volume and sur face area. Differential interference contrast (DIC, or Nomarski) imagi ng can enable cell edges to be clearly visualized in unstained tissue due to the slight difference in refractive index between aqueous media and cytoplasm. DIC is affected in only one direction-the direction of the optical shear. A 1-D edge detector was used in that direction wit h a scale length equal to that of an in-focus edge to highlight cell b oundaries. By comparison with the signal from the edge detector on an out-of-focus slice, the in-focus slices could be segmented and, after noise suppression, cell outlines obtained. A voxel paradigm was used t o calculate cell, volume and differential geometry was used for surfac e area estimation. We applied this approach to obtain 3-D dimensional information by optical sectioning of motile Amoeba proteus.