Andrew H. Song

SCOPE improves AI triage in 3D pathology by using morphology-aware prototypes to identify high-risk 2D slices for pathologist review.

SCALE3D improves prostate cancer risk prediction by analyzing 3D cell interactions through supercell graphs, outperforming traditional features and reducing computational demands.

This paper develops a deep learning method to virtually stain 3D microCT lung tissue, mimicking histological stains and enabling 3D pathology analysis.

CARP3D uses deep learning to prioritize high-risk 2D sections in 3D pathology data, improving disease detection while reducing pathologist workload.

This paper introduces TRACE, a deep learning model for automated liver histopathology assessment in preclinical drug safety, achieving higher concordance than pathologists in lesion scoring and dose-response analysis.

This paper introduces GEESE, an AI model that imputes morphomolecular signatures in toxicology data, linking gene expression changes to drug-induced liver injuries.