---
license: gpl-3.0
---

# Pre-trained Deep Learning Segmentation Models for ACE Pipeline  

## AI-based Cartography of Ensembles (ACE) Pipeline Highlights  

- Cutting-edge vision transformer and CNN-based deep learning architectures trained on large LSFM datasets to map brain-wide local/laminar neuronal activity.  
- Optimized cluster-wise statistical analysis with a threshold-free enhancement approach to chart subpopulation-specific effects at the laminar and local levels, without restricting the analysis to atlas-defined regions.  
- Modules for providing deep learning model uncertainty estimates and fine-tuning.  
- Interface with [MIRACL](https://miracl.readthedocs.io/en/latest/index.html) registration.  
- Ability to map the connectivity between clusters of activations.  

📖 **Read the full article here:** [LINK](https://www.nature.com/articles/s41592-024-02583-1)  

🔗 **MIRACL Software:** [LINK](https://miracl.readthedocs.io/en/latest/index.html)  

🔗 **ACE Installation Page:** [LINK](https://miracl.readthedocs.io/en/latest/tutorials/workflows/ace_flow/ace_flow.html)