CellBridge: EPFL trains H&E image models to identify cells using spatial transcriptomics
mo_lotfollahi · x · 2026-10-26
EPFL researchers released the CellBridge preprint. H&E slides reveal tissue architecture and cell appearance but often can't distinguish cell types (e.g., T vs B lymphocytes) without extra markers, while spatial transcriptomics provides molecular detail but is much harder to obtain routinely. CellBridge trains an image model on 1 million cells with paired H&E images and Xenium spatial transcriptomics, using molecular measurements to teach visual recognition of cell identity.
More from Research
- CoRL paper: flow-matching VLAs can run 1-step Euler inference with no retraining — kaixhin · 2026-10-26
- CoVer at EMNLP 2026: Training Code LLMs to Generate and Self-Verify Simultaneously — algo_diver · 2026-10-26
- Robotic dexterity should be judged on speed, force control and tool use, not copying humans — DJiafei · 2026-10-26
- Turing-complete malware can run in PyTorch's backward pass via a zero-valued loss — retr0jirachi · 2026-10-26
- Nathan Lambert's first open models lecture: distillation, risks and the ecosystem explained — natolambert · 2026-10-26
- Search researcher on MRL debate: losses are tolerable but benchmarks are too small with stale QRELs — srchvrs · 2026-10-26