Automatic Contour Extraction from 2D Neuron Images

The current work describes a novel system devised for automatic contour extraction of 2D branching structures images obtained from 3D neurons. Most contour-based methods for neuronal cell shape analysis fall short of suitable representation of such cells because overlaps between neuronal processes prevent traditional contour following algorithms from entering the innermost cell regions. The herein-proposed framework is specifically aimed at the problem of contour following even in presence of multiple overlaps. First, the input image is preprocessed in order to obtain an 8-connected skeleton with one-pixel-wide branches, as well as a set of subtree seed pixels and critical regions (i.e., bifurcations and crossings). Next, for each subtree, the tracking algorithm iteratively labels all valid pixel branches, up to a critical region, where the algorithm determines the suitable direction to proceed. Our algorithm has been found to exhibit robustness even for images with close parallel segments. Experimental results using real data (neural cell images) are presented.