New AI model generates gene expression profiles from medical images and data

Researchers have developed M$^3$-Gen, a novel framework designed to generate gene expression profiles using a Generative Adversarial Network. This model conditions the generation process on histopathology images and clinical metadata, learning a unified latent representation through contrastive learning. M$^3$-Gen produces biologically coherent gene expression data, as demonstrated on the TCGA dataset, and offers intrinsic explainability by identifying which image regions influenced specific gene expression outputs. AI

IMPACT This research could enable more cost-effective and privacy-preserving multimodal biomedical research by generating synthetic gene expression data.

RANK_REASON The cluster contains an academic paper detailing a new AI model and its methodology. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• arXiv
• Carlo Sgaravatti
• Generative Adversarial Network
• Hugging Face
• M$^3$-Gen

• paper
• model release

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →