Search papers, labs, and topics across Lattice.
This paper introduces information gain (IG) as a cell-wise evaluation metric for generative virtual staining (VS) models in high-throughput screening (HTS). IG allows for direct assessment of predicted posterior distributions, addressing the limitation of existing metrics that only evaluate marginal distribution accuracy. The authors demonstrate that IG reveals performance differences between diffusion- and GAN-based VS models on an HTS dataset that are not captured by other evaluation metrics.
Information gain offers a theoretically sound and interpretable way to directly assess the quality of predicted posterior distributions in virtual staining, revealing performance differences missed by traditional evaluation metrics.
Generative virtual staining (VS) models for high-throughput screening (HTS) can provide an estimated posterior distribution of possible biological feature values for each input and cell. However, when evaluating a VS model, the true posterior is unavailable. Existing evaluation protocols only check the accuracy of the marginal distribution over the dataset rather than the predicted posteriors. We introduce information gain (IG) as a cell-wise evaluation framework that enables direct assessment of predicted posteriors. IG is a strictly proper scoring rule and comes with a sound theoretical motivation allowing for interpretability, and for comparing results across models and features. We evaluate diffusion- and GAN-based models on an extensive HTS dataset using IG and other metrics and show that IG can reveal substantial performance differences other metrics cannot.
