Abstract
The paper proposes a way to examine whether cervical cells qualify as normal or abnormal on the basis of phase profiles, and to subsequently perform local fractal descriptions of their three dimensional holographic images. The method is an alternative tool to discriminate among normal and abnormal cervical cells by exploiting the whole image of the cell not only limited zones like random cross sections. The empirical findings indicate correlation between cross sectional phase profile and fractal characterisation. The results are useful for developing cell counting algorithms based on Bayesian programming.