Xubo B. Song

MiroSCOPE is an AI platform for annotating functional tissue units in digital pathology, enabling efficient annotation of over 71,900 units across prostate cancer samples.

This study aggregates multimodal cancer data across unaligned embedding spaces and finds that tumor of origin signal is preserved for learning tasks.

This paper introduces Graph Structured Neural Networks (GSNN) for perturbation biology, showing they outperform baselines in predicting perturbed gene expression, drug combinations, and disease-specific drug prioritization.

The paper compares deep learning models for segmenting nuclei and nucleoli in cancer EM images using sparse labels and semi-supervised learning, showing improvements in 3D Dice scores.

This study improves model generalizability in biological object detection using electron microscopy images through normalization and augmentation techniques, achieving a 29% average precision boost.