Research map / Image generation / Autoregressive and latent generation
From Spots to Pixels: Dense Spatial Gene Expression Prediction from Histology Images
Ruikun Zhang, Yan Yang, Liyuan Pan
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 |
|---|---|---|---|
| replace SAFB decoder with plain Conv (ablation) | PixNet · Breast Visium HD (ablation) | PCC@M: 0.169 -> 0.325 | Improved |
| replace SAFB decoder with ViT (ablation) | PixNet · Breast Visium HD (ablation) | PCC@M: 0.188 -> 0.325 | Improved |
| add PCC loss to MSE loss (ablation) | PixNet · Breast Visium HD (ablation) | PCC@M: 0.293 -> 0.325 | Improved |
| add MSE loss to PCC loss (ablation) | PixNet · Breast Visium HD (ablation) | MSE: 0.192 -> 0.153 | Improved |
| train with 2um spots in addition to 16+8um (ablation) | PixNet · Breast Visium HD (ablation) | PCC@M: 0.303 -> 0.325 | Improved |
| train with all spot sizes vs only 16um (ablation) | PixNet · Breast Visium HD (ablation) | PCC@M: 0.244 -> 0.325 | Improved |
| dense prediction vs fixed-spot methods, trained at 100um test at 2um | PixNet · Visium HD 2um (train STNet) | PCC@M: 0.118 (SGN) -> 0.198 | Improved |
Closest papers by meaning
- Preserving DEG Rankings for Gene Discovery in Histology-Based Spatial Gene Expression Prediction (NeurIPS 2026 · similarity 0.90)
- MERGE: Multi-faceted Hierarchical Graph-based GNN for Gene Expression Prediction from Whole Slide Histopathology Images (CVPR 2025 · similarity 0.90)
- Diffusion Generative Modeling for Spatially Resolved Gene Expression Inference from Histology Images (ICLR 2025 · similarity 0.90)
- Scalable Generation of Spatial Transcriptomics from Histology Images via Whole-Slide Flow Matching (ICML 2025 · similarity 0.90)
- STimage-1K4M: A histopathology image-gene expression dataset for spatial transcriptomics (NeurIPS 2024 · similarity 0.89)
- M2OST: Many-to-one Regression for Predicting Spatial Transcriptomics from Digital Pathology Images (AAAI 2025 · similarity 0.89)
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.
