Research map / Vision-language and multimodal / Cross-modal learning
Fusing Pixels and Genes: Spatially-Aware Learning in Computational Pathology
Minghao Han, Dingkang Yang, Linhao Qu, Zizhi Chen, Gang Li, Han Wang, Jiacong Wang, Lihua Zhang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| fine-tune PLIP with gene expression supervision | PLIP · DLPFC linear probing | balanced accuracy: 0.429 -> 0.476 | Improved |
| fine-tune CONCH with gene expression supervision | CONCH · DLPFC linear probing | balanced accuracy: 0.454 -> 0.481 | Improved |
| fine-tune CONCH with gene expression supervision | CONCH · HBC linear probing | balanced accuracy: 0.704 -> 0.811 | Improved |
| add spatial-aware CGR loss to gene encoder pretraining (ablation) | STAMP_G · gene encoder pretraining | linear probing / clustering | Improved |
| use only IGR loss for gene encoder (ablation) | STAMP_G · gene encoder pretraining | linear probing / clustering | Hurt |
| add vanilla contrastive alignment of UNI with STAMP_G (L_P-S) (ablation) | STAMP_V · DLPFC | linear probing | Improved |
| add region-gene loss L_R-S (ablation) | STAMP · DLPFC and PSC | downstream tasks | Improved |
| add patch-region intra-modal loss L_P-R (ablation) | STAMP · DLPFC and PSC | downstream tasks | Improved |
| add cross-scale patch positioning loss L_CSP (ablation) | STAMP · DLPFC and PSC | downstream tasks | Improved |
| use SpaVis-6M and stage 1 pretraining (ablation) | STAMP · DLPFC and HBC | linear probing and clustering | Improved |
| add alignment to gene encoder (STAMP_G dagger -> STAMP_G) (ablation) | STAMP_G · DLPFC linear probing | balanced accuracy: 0.571 -> 0.658 | Improved |
| fuse gene and vision embeddings (STAMP_V -> STAMP_F) | STAMP · DLPFC linear probing | balanced accuracy: 0.624 -> 0.721 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- mSTAR — MSTAR: Box-free Multi-query Scene Text Retrieval with Attention Recycling (NeurIPS 2025) · better on 4 of 4 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- M2OST: Many-to-one Regression for Predicting Spatial Transcriptomics from Digital Pathology Images (AAAI 2025 · similarity 0.91)
- Diffusion Generative Modeling for Spatially Resolved Gene Expression Inference from Histology Images (ICLR 2025 · similarity 0.90)
- Multi-modal Topology-embedded Graph Learning for Spatially Resolved Genes Prediction from Pathology Images with Prior Gene Similarity Information (CVPR 2025 · similarity 0.89)
- STimage-1K4M: A histopathology image-gene expression dataset for spatial transcriptomics (NeurIPS 2024 · similarity 0.89)
- HEST-1k: A Dataset For Spatial Transcriptomics and Histology Image Analysis (NeurIPS 2024 · similarity 0.88)
- Towards Scalable Context-Aware Single-Cell Spatial Transcriptomics Prediction from Histology Images (NeurIPS 2026 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
