Red Blood Cell Surface Segmentation Based on Shape Reconstruction and Statistics Feature Extraction

There is considerable evidence demonstrating that erythrocyte deformability is playing an important role in the filerability of blood and consequently in the pathology of a variety of blood-related diseases. The shape of red blood cell is also a determining factor for its deformability and filerability. The original input red blood cell image we are aiming to deal with is captured by Scanned Electronic Microscope (SEM). We map the intensity level image into a 3-D depth through Shape from Shading (SFS) first to get the height field of each pixel, then each cell on top level is extracted from the whole image by means of region growing algorithm based on boundary contour tracing, which are used for feature extraction and statistics computation. The experimental result shows that this approach is easily to implement and promising.