Related Experiment Video
Updated: May 26, 2026
10:22
Comprehensive Spatial Profiling of Species-agnostic Transcriptomes via Stereo-seq
Published on: October 31, 2025
A comprehensive survey of computer vision methods for spatial transcriptomics
Junchao Zhu1, Ruining Deng2, Junlin Guo3
1Department of Computer Science, Vanderbilt University, 2301 Vanderbilt Place, Nashville, TN 37235, USA.
Briefings in Bioinformatics
|May 25, 2026
Summary
Computer vision AI enhances spatial transcriptomics (ST) by integrating tissue morphology with gene expression data. This approach reduces costs, enables virtual sequencing, and advances 3D spatial omics analysis for research and clinical applications.
Area of Science:
- Computational Biology
- Bioinformatics
- Artificial Intelligence
Background:
- Spatial transcriptomics (ST) measures gene expression and spatial localization in tissues.
- Current bioinformatics approaches in ST often underutilize detailed morphological information.
- Limitations of ST include high cost, limited clinical use, and 2D analysis of 3D tissues.
Purpose of the Study:
- To survey computer vision AI models applied to spatial transcriptomics analytics.
- To explore how AI can integrate histological morphology with molecular data in ST.
- To address key limitations of ST, such as cost and clinical applicability.
Main Methods:
- Systematic survey and categorization of computer vision AI models for ST.
- Analysis of approaches based on architectures, learning paradigms, tasks, and datasets.
- Tracing the technological evolution of vision-driven ST.
Main Results:
- Computer vision AI offers novel approaches beyond conventional bioinformatics for ST.
- AI models can predict ST from histology (virtual sequencing), reducing costs.
- AI enables 3D reconstruction of tissue models for advanced spatial omics analysis.
Conclusions:
- Computer vision AI significantly enhances spatial transcriptomics analytics.
- Vision-driven ST has the potential to transform basic research and clinical practice.
- This survey provides a panoramic perspective on the field and future directions.
