The research of microscopic image segmentation and recognition on the cancer cells fallen into peritoneal effusion

Auto-segmentation of cells is one of the most interesting segmentation problems due to the complex nature of the cell tissues and to the inherent problems of video microscopic images. Objects, which are variant, narrow range of gray levels, non-random noise, are ubiquitous problems presented in this kind of image. Considering the above characteristics, an adaptive min-distance algorithm is proposed in this paper, which is available to segment the suspected cell and nucleus from the complex background in the microscopic image of cells fallen into peritoneal effusion. 15 features of the cancer cell and calculating formulas are presented respectively. These features are employed to construct a backpropagation neural network classifier which classifies and recognizes the cancer cells fallen into peritoneal effusion. Tests are performed using clinical cases recommended by the pathologists, results show that the proposed algorithm can efficiently segment the cell image and receive higher accuracy of cancer cell diagnosis