Characterizing the Features of Mitotic Figures Using a Conditional Diffusion Probabilistic Model
This provides a new interpretability tool for pathologists to understand and communicate features in mitosis detection, addressing human subjectivity in labeling, but it is incremental as it builds on existing diffusion models for a specific medical imaging task.
The paper tackled the problem of unreliable mitotic figure labels in histology images by training a conditional diffusion model to synthesize cell nuclei patches and generate sequences showing nuclei transitioning into mitotic states, identifying features like cytoplasm granularity and nuclear irregularity as key indicators.
Mitotic figure detection in histology images is a hard-to-define, yet clinically significant task, where labels are generated with pathologist interpretations and where there is no ``gold-standard'' independent ground-truth. However, it is well-established that these interpretation based labels are often unreliable, in part, due to differences in expertise levels and human subjectivity. In this paper, our goal is to shed light on the inherent uncertainty of mitosis labels and characterize the mitotic figure classification task in a human interpretable manner. We train a probabilistic diffusion model to synthesize patches of cell nuclei for a given mitosis label condition. Using this model, we can then generate a sequence of synthetic images that correspond to the same nucleus transitioning into the mitotic state. This allows us to identify different image features associated with mitosis, such as cytoplasm granularity, nuclear density, nuclear irregularity and high contrast between the nucleus and the cell body. Our approach offers a new tool for pathologists to interpret and communicate the features driving the decision to recognize a mitotic figure.