{"ID":2884076,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2508.07225","arxiv_id":"2508.07225","title":"HaDM-ST: Histology-Assisted Differential Modeling for Spatial Transcriptomics Generation","abstract":"Spatial transcriptomics (ST) reveals spatial heterogeneity of gene expression, yet its resolution is limited by current platforms. Recent methods enhance resolution via H\u0026E-stained histology, but three major challenges persist: (1) isolating expression-relevant features from visually complex H\u0026E images; (2) achieving spatially precise multimodal alignment in diffusion-based frameworks; and (3) modeling gene-specific variation across expression channels. We propose HaDM-ST (Histology-assisted Differential Modeling for ST Generation), a high-resolution ST generation framework conditioned on H\u0026E images and low-resolution ST. HaDM-ST includes: (i) a semantic distillation network to extract predictive cues from H\u0026E; (ii) a spatial alignment module enforcing pixel-wise correspondence with low-resolution ST; and (iii) a channel-aware adversarial learner for fine-grained gene-level modeling. Experiments on 200 genes across diverse tissues and species show HaDM-ST consistently outperforms prior methods, enhancing spatial fidelity and gene-level coherence in high-resolution ST predictions.","short_abstract":"Spatial transcriptomics (ST) reveals spatial heterogeneity of gene expression, yet its resolution is limited by current platforms. Recent methods enhance resolution via H\u0026E-stained histology, but three major challenges persist: (1) isolating expression-relevant features from visually complex H\u0026E images; (2) achieving s...","url_abs":"https://arxiv.org/abs/2508.07225","url_pdf":"https://arxiv.org/pdf/2508.07225v1","authors":"[\"Xuepeng Liu\",\"Zheng Jiang\",\"Pinan Zhu\",\"Hanyu Liu\",\"Chao Li\"]","published":"2025-08-10T08:09:06Z","proceeding":"eess.IV","tasks":"[\"eess.IV\",\"cs.CV\",\"q-bio.QM\"]","methods":"[\"Diffusion Model\"]","has_code":false}