SLIM-Diff: Shared Latent Image-Mask Diffusion with Lp loss for Data-Scarce Epilepsy FLAIR MRI
This work addresses data scarcity in medical imaging for epilepsy diagnosis, but it is incremental as it builds on existing diffusion models with specific tuning for a domain-specific task.
The authors tackled the problem of generating joint image-mask pairs for scarce epilepsy FLAIR MRI data by proposing SLIM-Diff, a compact diffusion model with a shared-bottleneck U-Net and tunable Lp loss, resulting in improved image fidelity with L1.5 loss and better lesion mask preservation with L2 loss.
Focal cortical dysplasia (FCD) lesions in epilepsy FLAIR MRI are subtle and scarce, making joint image--mask generative modeling prone to instability and memorization. We propose SLIM-Diff, a compact joint diffusion model whose main contributions are (i) a single shared-bottleneck U-Net that enforces tight coupling between anatomy and lesion geometry from a 2-channel image+mask representation, and (ii) loss-geometry tuning via a tunable $L_p$ objective. As an internal baseline, we include the canonical DDPM-style objective ($ε$-prediction with $L_2$ loss) and isolate the effect of prediction parameterization and $L_p$ geometry under a matched setup. Experiments show that $x_0$-prediction is consistently the strongest choice for joint synthesis, and that fractional sub-quadratic penalties ($L_{1.5}$) improve image fidelity while $L_2$ better preserves lesion mask morphology. Our code and model weights are available in https://github.com/MarioPasc/slim-diff