Automatic Cell Cycle Localization Using Latent-Dynamic Conditional Random Fields

This paper proposes an automatic cell cycle localization method based on the Latent-Dynamic Conditional Random Fields (LDCRFs) model. Since the LDCRFs model can jointly capture both extrinsic dynamics and intrinsic sub-structure, it can simultaneously model the visual dynamics within one stage and visual transition between adjacent stages in one mitosis sequence. Based on our previous work on candidate mitosis sequence extraction and classification, this paper mainly focuses on the formulation of LDCRFs for cell cycle modeling. Besides, the model learning and inference methods are also presented. The evaluation on C2C12 dataset shows the superiority of the proposed method.