AI RESEARCH
Latent Causal Modeling for 3D Brain MRI Counterfactuals
arXiv CS.AI
•
ArXi:2409.05585v3 Announce Type: replace-cross The number of samples in structural brain MRI studies is often too small to properly train deep learning models. Generative models show promise in addressing this issue by effectively learning the data distribution and generating high-fidelity MRI. However, they struggle to produce diverse, high-quality data outside the distribution defined by the