Automatic Liver Tissue Section Image Characterization

Analyzing liver cross-section images is a manual and laborious task, slowing down critical research toward finding alternative cures for patients with end-stage liver disease. In this paper, we present methods to automatically count hepatocytes cells, count nuclei and classify liver vessel types, using input images of cell boundary and cell nuclei based on a dataset of 21 images. Compared to a trained researcher, the methods are able count cells, segment overlapping nuclei, and classify vessel types, including portal vein, central vein, and bile duct with reasonable precision and accuracy. Future work includes improving classification accuracy and detecting other cell types.