Nevenka Dimitrova

Beyond Segmentation Accuracy: Evaluating Biological Fidelity in Spatial Pathology

The paper evaluates how segmentation accuracy affects biological fidelity in spatial pathology, finding that segmentation errors significantly impact tumor-immune interactions despite similar segmentation performance.

Discovering Latent Scientific Concepts through Discrete Representation Learning

This paper introduces Kodiak, a discrete representation learning method for scientific imaging that improves representation quality and enables interpretable, reusable concepts across domains.

Metastatic site influences driver gene function in pancreatic cancer

This study shows that metastatic site influences the function of driver genes like SMAD4 in pancreatic cancer, with organ-specific epigenetic states affecting tumor growth.