Agentic data science with the Data Agent Kit
cloud.google.com·7h ago
AI Research
arXiv CS.AI·Aug 11·5 min read
Decoding Phenotypes: A Framework for Fusing Genomic Language Models and Neuroimaging
TL;DR
GeneFuse is a novel multimodal learning framework that integrates genomic data from Genomic Language Models with neuroimaging features to enhance disease diagnosis. It outperforms existing methods in identifying cognitive decline and dementia, achieving AUROCs of 0.77 and 0.83.
✦ Why It Matters
Consider integrating GeneFuse into your diagnostic workflows to improve accuracy in identifying cognitive decline.
Key Takeaways
- Explore GeneFuse for multimodal data integration in diagnostics.
- Utilize Genomic Language Models to enhance genetic data representation.
- Implement GCFM to modulate imaging features with genomic context.
- Adopt U-GRF to account for uncertainty in predictive modeling.
How It Works
GeneFuse combines genomic data and neuroimaging by using GCFM to adjust image features based on genetic embeddings, allowing for a more nuanced understanding of disease. U-GRF further refines this integration by incorporating uncertainty from imaging predictions, ensuring that the most relevant genetic information is emphasized in the diagnostic process.