Zhejiang University School of Medicine

Cancer-registry encoder pretrained by masked field-value modelling on 9.4M SEER tumour records, carrying survival signal into unseen rare cancers.

Renal tumor histopathology model that detects tissue regions, classifies nine subtypes, grades nuclei and scores prognosis from a single H&E slide.

Multi-modal transformer fusing LLM gene embeddings with biological knowledge graphs to predict single-cell responses to genetic perturbations.

Virtual staining from H&E histology to immunohistochemistry, using a Schrödinger bridge trained on unpaired slides with a topology-preserving term.

Vision-language foundation model for kidney cancer CT, covering zero-shot malignancy diagnosis, report generation, and recurrence risk prediction.

Genome-anchored histopathology embeddings that predict molecular biomarkers, subtypes, and survival from whole-slide images alone at inference.

CLIP-based vision-language foundation model for eye imaging, enabling zero-shot disease detection and cross-modal retrieval across 11 modalities.

Multimodal drug-response model coupling cell and molecule foundation models, pretrained on 1.8M perturbation RNA-seq profiles over 22,000 compounds.

Cross-domain molecular foundation model encoding small molecules, protein pockets, and their complexes in 2D and 3D on one Transformer backbone.

RNA-binding protein binding profiles predicted base by base across 800 bp windows, with an m6A signal channel that exposes methylation-RBP crosstalk.

Cell-free RNA language model for multi-cancer detection, classifying plasma samples straight from raw sequencing reads without gene annotation.