SLIM-Diff：共享潜在图像-掩码扩散模型与Lp损失用于数据稀缺的癫痫FLAIR MRI
SLIM-Diff: Shared Latent Image-Mask Diffusion with Lp loss for Data-Scarce Epilepsy FLAIR MRI
AI总结：
SLIM-Diff通过共享瓶颈U-Net和L_p损失提升癫痫FLAIR MRI中稀少病变的联合生成效果。
摘要：
癫痫FLAIR MRI中的局灶性皮层发育不良病变微妙且稀少，使得联合图像-掩码生成模型容易出现不稳定和记忆现象。我们提出了SLIM-Diff，一种紧凑的联合扩散模型，其主要贡献是（i）一个单一的共享瓶颈U-Net，通过2通道图像+掩码表示强制解剖结构和病变几何形状之间的紧密耦合，以及（ii）通过可调的L_p目标进行损失-几何调节。作为内部基准，我们包括了经典的DDPM式目标（ε预测与L2损失）并在此匹配设置下隔离预测参数化和L_p几何的影响。实验表明，x0预测始终是联合合成的最佳选择，而分数次二次惩罚（L1.5）提高了图像保真度，而L2损失更好地保持了病变掩码的形态。我们的代码和模型权重可在https://github.com/MarioPasc/slim-diff上获得。
英文摘要：
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
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
