The field of spatial transcriptomics and digital pathology is rapidly advancing, with a focus on developing innovative methods for analyzing and integrating multimodal data. Recent studies have highlighted the importance of incorporating biological semantics and spatial context into computational models, enabling a deeper understanding of tissue microenvironments and cellular heterogeneity. Noteworthy papers have introduced novel frameworks for spatial transcriptomics data clustering, multiscale integration of nuclear morphology and microenvironmental context, and adaptive multi-scale integration for robust cell annotation. Additionally, advancements in digital pathology have led to the development of unified models for digital hematopathology, slide-label aware multitask pretraining, and generalizable multiple instance learning frameworks. These innovations have the potential to revolutionize our understanding of human biology and disease, and will likely have a significant impact on the field in the coming years. Notable papers include SemST, which leverages Large Language Models to enable genes to 'speak' through their symbolic meanings, and HiFusion, which integrates hierarchical intra-spot alignment and regional context fusion for spatial gene expression prediction. GROVER is also noteworthy, as it proposes a novel framework for adaptive integration of spatial multi-omics data.
Advancements in Spatial Transcriptomics and Digital Pathology
Sources
When Genes Speak: A Semantic-Guided Framework for Spatially Resolved Transcriptomics Data Clustering
Latent space models for grouped multiplex networks
Multimodal Posterior Sampling-based Uncertainty in PD-L1 Segmentation from H&E Images
Multiscale Grassmann Manifolds for Single-Cell Data Analysis
GROVER: Graph-guided Representation of Omics and Vision with Expert Regulation for Adaptive Spatial Multi-omics Fusion
Rank-Aware Agglomeration of Foundation Models for Immunohistochemistry Image Cell Counting
Enhancing Neuro-Oncology Through Self-Assessing Deep Learning Models for Brain Tumor Unified Model for MRI Segmentation
HiFusion: Hierarchical Intra-Spot Alignment and Regional Context Fusion for Spatial Gene Expression Prediction from Histopathology
Edge-aware baselines for ogbn-proteins in PyTorch Geometric: species-wise normalization, post-hoc calibration, and cost-accuracy trade-offs
Adaptive Multi-Scale Integration Unlocks Robust Cell Annotation in Histopathology Images
Mitigating Spurious Correlations in Patch-wise Tumor Classification on High-Resolution Multimodal Images
Tissue Aware Nuclei Detection and Classification Model for Histopathology Images
Rare Genomic Subtype Discovery from RNA-seq via Autoencoder Embeddings and Stability-Aware Clustering
Gene Incremental Learning for Single-Cell Transcriptomics
Uni-Hema: Unified Model for Digital Hematopathology
3D-Guided Scalable Flow Matching for Generating Volumetric Tissue Spatial Transcriptomics from Serial Histology
SLAM-AGS: Slide-Label Aware Multi-Task Pretraining Using Adaptive Gradient Surgery in Computational Cytology
nnMIL: A generalizable multiple instance learning framework for computational pathology
CellGenNet: A Knowledge-Distilled Framework for Robust Cell Segmentation in Cancer Tissues
Deep Pathomic Learning Defines Prognostic Subtypes and Molecular Drivers in Colorectal Cancer
SIGMMA: Hierarchical Graph-Based Multi-Scale Multi-modal Contrastive Alignment of Histopathology Image and Spatial Transcriptome