Multiresolution fractal analysis and classification of neurite images

Biological images are critically important for better understanding of the structure and functioning of cells and proteins. Automated image analysis of neuronal cells is essential for neuroscience research and is becoming a central component for quantifying the effect of candidate drugs on cells. To investigate the intricate nervous processes involved in many biological activities by computerized image analysis, accurate analysis and classification of neurites are prerequisite. In this paper, the fractal properties exhibited by neurons are further investigated and measures derived from multiresolution fractal analysis are exploited in differentiating neuron types by machine learning methods. The proposed method can serve as a candidate tool for large-scale neurite analysis.