ROBUST CELL-NUCLEI SEGMENTATION USING STATISTICAL MODELING

The objective analysis of cytological and histological images has been the subject of research for many years. One of the most difficult fie lds in histological image analysis is the automated segmentation of ti ssue-section images. We propose a multistage segmentation method for t he isolation of cell nuclei in such images. In the first stage the com pact Hough transform (CHT) is used to determine possible locations of the nuclei. We then define a likelihood function which enables us to p erform an optimization procedure based on the maximization of this fun ction. Global grey-level histogram information is used thoughout the a lgorithm to reduce the amount of computation and to reject unwanted ar tefacts. The algorithm is tested on connective tissue images with very encouraging results. Apart from detecting well-separated nuclei in th e images, it performs well in separating dividing nuclei into likely s ubstructures. At the same time the algorithm provides us with a measur e of uncertainty along the detected boundary, in the form of the value of the likelihood function.