Automated Segmentation of ELA Cancer Cells in Microscopic Images for Evaluating the Cytotoxic Effect of Selected Medicinal Plants

This paper proposes automated segmentation methods for dead cancer cells in the microscopic images of Ehrlich's Lymphoma Ascite (ELA) cancer cells of Swiss albino mice to evaluate the cytotoxic effect of the medicinal plants Cynodon dactylon (C. dactylon) and Terminalia catappa (T. catappa). Microscopic images of untreated ELA cancer cell samples and these treated with extracts of C. dactylon and T. catappa are acquired. The dead cancer cells in the microscopic images are identified and isolated using region-based mathematical morphology and watershed segmentation methods. The dead cancer cells in the output images are counted using run-length encoding method. The microscopic image of the untreated sample has only nine dead cancer cells whereas those of samples treated with the extracts each have 31 dead cancer cells. The percentage of dead ELA cancer cells in the samples increased by 244.45% after treatment with the extracts of C. dactylon and T. catappa, respectively. It is thus concluded that C. dactylon and T. catappa have a significant cytotoxic effect against ELA cancer cells.